Factors Affecting the Abundance of a Declining Grassland Bird

166 Pages·2016·2.21 MB·English

Factors Affecting the Abundance of a Declining Grassland Bird: Implications
for Recovery Strategy Planning and Implementation

by
Danielle Marie Ethier

A Thesis
presented to
The University of Guelph

In partial fulfillment of the requirements
for the Degree of
Doctor of Philosophy
in
Integrative Biology

Guelph, Ontario, Canada
© Danielle Marie Ethier, August 2016
ABSTRACT

FACTORS AFFECTING THE ABUNDANCE OF A DECLINING GRASSLAND BIRD:
IMPLICATIONS FOR RECOVERY STRATEGY PLANNING AND
IMPLEMENTATION

Danielle M. Ethier Advisors: Dr. Thomas Nudds
University of Guelph, 2016 Dr. Steve Crawford

Uncertainty is pervasive in all conservation decisions, the systematic treatment of which is
necessary when evaluating the causes of species endangerment and deciding on appropriate
actions to better ensure persistence. In Canada, a guiding principal of endangered species
legislation for reducing uncertainty is adaptive management; however, adaptive management is
not explicit in the development and application of most recovery strategies. To set the context for
adaptive management recovery programming for at-risk species in Ontario, Canada, I
synthesized existing knowledge, articulate and test critical uncertainties as hypotheses, and begin
model development to predict expected outcomes of management alternatives, using the geo-
politically defined population of Bobolink (Dolichonyx oryzivorus) in Ontario as a case study.
Specifically, in Chapter 1, I combine evidence from empirical studies and expert judgement to
model causal mechanisms driving Bobolink abundance dynamics and characterize the structural
complexity of the management system. In Chapter 2, I spatially resolved Bobolink abundance
trends to determine which areas of the province are contributing most substantially to overall
abundance decline, and which presently satisfy recovery targets. In Chapter 3, I examined spatio-
temporal variability in landscape pattern and processes to determine by which mechanism(s), and
to what extent, changes in quantity and quality of agricultural grassland habitats have contributed
to regional changes in abundance. Finally, in Chapter 4, I developed regionally scaled habitat
suitability index (HSI) models to facilitate development of predictive population viability
models. HSI models were used to evaluate to what extent lower-order proxies for processes of
habitat selection influence patterns in abundance regionally. My research highlights how a
science-based approach to recovery strategy planning can more effectively identify and address
uncertainties in knowledge, and be used as an avenue to set the foundation of an adaptive
management program.
iv

Dedication

In loving memory of my Old Man,
Richard L. Ethier
1945-2015
v

Acknowledgments
Foremost, I thank my husband, Josh Sayers, for his unwavering support throughout my
seemingly endless academic pursuits, and our daughter, Annica, who reminds me every day that
there is much more to life than research.
My deepest gratitude to my advisor Dr. Thomas Nudds, who agreed to take on one last
PhD student at the beginning of his retirement. He has taught me the importance of asking the
first question first, and being an advocate for science-based decision making for species-at-risk.
My co-advisor, Dr. Steve Crawford, offered timely advice and helpful comments to improve the
clarity of my thesis. My committee member, Dr. Nicola Koper, offered statistical guidance,
editorial support, and cheerful encouragement throughout my PhD.
I thank the past and current members of the Nudds, Norris, Newman, McAdam, McCann,
and Fryxell labs for their friendship and research support. I am especially grateful to my office
mates (Laura Trout, Rachael Derbyshire, Grace Pittman, Alex Sutton, and Samantha Knight) and
lab mates (Van La, Allan Derbertin, Zan Zhang, and Kevin Reid) for making every day at the
office fun and inspiring. I acknowledge the countless volunteers involved in collecting North
American Breeding Bird Survey and Ontario Breeding Bird Atlas data, including their official
sponsors. I thank the anonymous stakeholders that participated in the elicitation workshop and
online questionnaire. Mike Brienesse and Kevin Reid provided guidance on stakeholder
elicitation and development of Bayesian Belief Networks.
This research was funded by the Ontario Ministry of Natural Resources and Forestry,
Natural Sciences and Engineering Research Council of Canada (NSERC) post-graduate
scholarship, University of Guelph Dean’s Tri-council Scholarship, and an I.O.D.E. War
Memorial Doctorate Scholarship.
vi

Table of Contents

Abstract ........................................................................................................................................... ii
Acknowledgments........................................................................................................................... v
List of Tables ................................................................................................................................. ix
List of Figures ................................................................................................................................. x
Prologue .......................................................................................................................................... 1
Thesis Goal .................................................................................................................................. 3
Background ................................................................................................................................. 5
Agricultural Grasslands ........................................................................................................... 5
Grassland Birds........................................................................................................................ 6
Bobolink (Dolichonyx oryzivorus) .......................................................................................... 6
Endangered Species Act (2007) .............................................................................................. 8
Exemption Regulation ............................................................................................................. 9
Recovery Goals, Objectives and Actions ................................................................................ 9
Literature Cited ......................................................................................................................... 11
Figures ....................................................................................................................................... 15
Co-authorship and Chapter Publication Status ............................................................................. 16
Chapter 1: Complexity of Factors Affecting Bobolink Population Dynamics Communicated with
Directed Acyclic Graphs ............................................................................................................... 17
Abstract ..................................................................................................................................... 17
Introduction ............................................................................................................................... 18
Methods ..................................................................................................................................... 20
Results ....................................................................................................................................... 21
1. Habitat Quantity ............................................................................................................. 21
1.1 Breeding habitat quantity ......................................................................................... 22
1.2 Non-breeding habitat quantity .................................................................................. 23
2. Habitat quality ................................................................................................................ 24
2.1 Breeding habitat quality ........................................................................................... 24
2.1.1 Landscape condition .......................................................................................... 24
2.1.2 Patch condition .................................................................................................. 25
2.1.3 Social information ............................................................................................. 26
2.2 Non-breeding habitat quality .................................................................................... 27
vii

3. Fecundity........................................................................................................................ 27
3.1 First hay harvest date................................................................................................ 28
3.2 Livestock trampling.................................................................................................. 29
3.3 Nest predation and parasitism .................................................................................. 30
3.4 Pesticides .................................................................................................................. 31
3.5 Prey availability........................................................................................................ 31
4. Mortality ........................................................................................................................ 32
4.1 Breeding period mortality ........................................................................................ 32
4.1.1 Date of first harvest ........................................................................................... 33
4.1.2 Pesticides ........................................................................................................... 33
4.1.3 Collisions with tall structures ............................................................................ 33
4.2 Non-breeding period mortality ................................................................................. 33
4.2.1 Pesticides ........................................................................................................... 34
4.2.2 Persecution......................................................................................................... 34
States of Published and Stakeholder Knowledge ...................................................................... 34
Discussion ................................................................................................................................. 35
Tables ........................................................................................................................................ 48
Figures ....................................................................................................................................... 49
Chapter 2: Scalar considerations in population abundance trend estimates: implications for
recovery strategy planning for species of conservation concern .................................................. 53
Abstract ..................................................................................................................................... 53
Introduction ............................................................................................................................... 54
Methods ..................................................................................................................................... 57
Study area........................................................................................................................... 57
Spatially explicit trend estimation ..................................................................................... 57
Results ....................................................................................................................................... 61
Discussion ................................................................................................................................. 62
Literature cited .......................................................................................................................... 67
Figures ....................................................................................................................................... 71
Chapter 3: Spatio-temporal variation in mechanisms driving regional-scale abundance patterns of
Bobolink (Dolichonyx oryzivorous) .............................................................................................. 74
Abstract ..................................................................................................................................... 74
Introduction ............................................................................................................................... 76
viii

Methods ..................................................................................................................................... 78
Study area........................................................................................................................... 78
Response variable .............................................................................................................. 78
Predictor variables ............................................................................................................. 80
Analysis.............................................................................................................................. 82
Results ....................................................................................................................................... 84
Spatio-temporal Bobolink trend estimates ......................................................................... 84
Spatio-temporal landscape change ..................................................................................... 85
Bayesian variable selection ................................................................................................ 86
Discussion ................................................................................................................................. 87
Literature cited .......................................................................................................................... 93
Tables ........................................................................................................................................ 98
Figures ..................................................................................................................................... 100
Chapter 4: Local habitat selection by grassland birds scales to regional extents: implications for
landscape-level conservation planning ....................................................................................... 105
Abstract ................................................................................................................................... 105
Introduction ............................................................................................................................. 106
Methods ................................................................................................................................... 108
Results ..................................................................................................................................... 112
Discussion ............................................................................................................................... 112
Literature cited ........................................................................................................................ 117
Figures ..................................................................................................................................... 121
Epilogue ...................................................................................................................................... 123
Literature cited ........................................................................................................................ 127
Supplemental Material ................................................................................................................ 128
Appendix A (Chapter 1).............................................................................................................. 128
Appendix B (Chapter 2) .............................................................................................................. 131
Appendix C (Chapter 2) .............................................................................................................. 135
Appendix D (Chapter 3).............................................................................................................. 136
Appendix E (Chapter 4) .............................................................................................................. 143
ix

List of Tables

Chapter 1

Table 1: Eight vegetation variables were identified as important predictors of Bobolink
(Dolichonyx oryzivorus)................................................................................................................ 48

Chapter 3

Table 1: Agricultural census divisions in Ontario, Canada which display strong evidence of
Bobolink (Dolichonyx oryzivorous) population growth or decline over the entire study period
(1986-2011) and five sub-periods.. ............................................................................................. 988
x

List of Figures

Prologue

Figure 1: Relative abundance and distribution map for Bobolink (Dolichonyx oryzivorous) based
on North American Breeding Bird Survey (BBS) data from 1987-2006 (Blancher 2009).. ........ 15

Chapter 1

Figure 1: Schematic diagram of a conceptual two-season life cycle model for Bobolink (adapted
from Pulliam and Danielson 1991). . ............................................................................................ 49

Figure 2: Directed acyclic graphs (DAGs) showing the causal relationships (links and arrows)
among variables (ovals) hypothesized to be driving changes in Bobolink habitat quantity on the
breeding and non-breeding grounds based on stakeholder elicitation and empirical evidence from
the literature.. ................................................................................................................................ 52

Figure 3: Directed acyclic graphs (DAGs) showing the causal relationships (links and arrows)
among variables (ovals) hypothesized to be driving changes in Bobolink habitat quality on the
breeding and non-breeding grounds based on stakeholder elicitation and empirical evidence from
the literature.. ................................................................................................................................ 53

Figure 4: Directed acyclic graphs (DAGs) showing the causal relationships (links and arrows)
among variables (ovals) hypothesized to be driving changes in Bobolink fecundity and mortality
on the breeding and non-breeding grounds based on stakeholder elicitation and empirical
evidence from the literature.. ........................................................................................................ 54

Chapter 2

Figure 1: Relative abundance and distribution map for Bobolink (Dolichonyx oryzivorous) based
on North American Breeding Bird Survey (BBS) data from 1987-2006 (Blancher 2009).. ........ 71

Figure 2: Posterior mean population trend estimates for Bobolink (Dolichonyx oryzivorous) from
A. 1967-2011 and B. 1998-2011 mapped to agricultural census divisions in southern Ontario,
Canada, using spatially explicit hierarchical models in a Bayesian statistical framework.. ......... 72

Figure 3: Box plots of posterior mean trends (solid centre line in box) for Bobolink (Dolichonyx
oryzivorous) from A. 1967-2011 and B. 1998-2011 shown across 35 agricultural census divisions
in Ontario, Canada (ordered from west to east).. .......................................................................... 73

Chapter 3

Figure 1: Directed acyclic graph showing causal relationships (link and arrows) among variables
(ovals) hypothesized to account for population declines in Bobolink (Dolichonyx oryzivorous)
(diamond) based on evidence in the published literature and/or stakeholder consultation. ....... 100