i

MANAGEMENT AND LANDSCAPE EFFECTS ON BENEFICIAL TALLGRASS

PRAIRIE INSECTS – A STUDY OF BEES AND BEETLES

BY

REID B. MILLER

A thesis submitted to the Faculty of Graduate Studies of the University of Manitoba in partial

fulfilment of the requirements of the degree of

MASTER OF SCIENCE

Department of Entomology

University of Manitoba

Winnipeg, Manitoba

Copyright © 2021 by Reid B. Miller

ii

iii

ABSTRACT

The largest remnant tall grass prairie (TGP) in Canada is the Manitoba Tall Grass Prairie

Preserve, located in the Rural Municipality of Stuartburn in southeastern Manitoba. TGP is home

to several rare and endangered plants and animals, particularly insects, whose contributions to

prairie functioning too often go unnoticed. Beneficial insects in TGP include pollinating insects,

such as wild bees, predatory insects, and insects that aid in decomposition/nutrient cycling.

Efforts by prairie managers to maintain remnant prairie include planned disturbance via

controlled fires and cattle grazing, with an aim to prevent woody encroachment and maximize

prairie plant diversity. Unfortunately, the effects of these practices on beneficial prairie insects

are often not considered. Because of this, I decided to collect beneficial insects in prairie sites

that differed in the type of disturbance that they were exposed to, the focal and surrounding

landcover of prairie patches, and local ground cover. Wild bees were caught in bee bowls over

the course of two growing seasons, and with blue vane traps in the second growing season.

Haplogastran beetles, comprising rove beetles, scarab beetles, and their relatives, were caught in

baited pitfall traps through two growing seasons. Once the bees and beetles were identified,

analysis via generalized linear mixed effect models, rarefaction curves, and distance-based

redundancy analysis uncovered trends in how these groups, as well as within group nesting

(bees) and feeding (beetles) guilds differed by disturbance type, landscape context, and local

ground cover. Wild bees appear to be negatively affected by cattle grazing, whereas beetles,

especially decomposer beetles, show greatest abundance in grazed sites. Ground nesting bees and

predatory beetles were both more abundant on burned sites compared to sites that had been

grazed. Landscape context also appears to affect bee and beetle abundance and diversity, with

bees demonstrating greater abundance, richness, and diversity under greater landscape diversity,

and both bees and beetles demonstrating increased abundance and diversity with increased forest

and ditch cover in the landscape. Ground nesting bees and predatory beetles showed opposite

responses to local ground cover, with bees favouring increased forb and bare ground cover, and

predatory beetles favouring increased grass cover. This study recovered several insect species

that were previously unknown from the province, and at least one previously

uncaptured/undescribed beetle species. Finding such diversity in such a narrow span of habitat

highlights the need to further consider the effects of disturbance and landscape context on the

beneficial insects that rely on the threatened TGP. It is hoped that this study can better inform

iv

land managers and property owners on the best ways to conserve the hidden and fragile insect

diversity that still exists in remnant Manitoba prairies.

v

ACKNOWLEDGMENTS

First and foremost, I must thank Jason for his help, feedback, and understanding. I also

owe a lot of gratitude to the rest of the Gibbs and Bobiwash labs for their feedback and

friendship. I am grateful for all the fieldwork help provided by Jason, Dave Holder, Olivia Lano,

Joel Gardner, Gina Karam, Bob Miller, Angela Reimer, and especially Alanna Shaw. For help in

the lab, I wish to thank Leah Irwin, Kathy Morgan, Amber Bass, Alanna Shaw, Syd Shukla -

Bergen, Joel Gardner, and Carl Dizon. Many thanks go to Dave Holder again for project design

and construction.

Several students in the department have been instrumental in supporting me throughout

this most formative part of my life, including Megan Colwell, Crystal Almdal, Gina Karam,

Emily Hanuschuk, Mike Killewald, and Massimo Martini, and all the other Entomology grad

students past and present. I am also very indebted to many of the Entomology/Animal Science

support staff, especially Kathy Graham, and the late Teresa Sitko, who was always very

welcoming on my bright and early starts.

I am also very privileged to have been able to interact and learn from all the Entomology

professors and instructors who have helped guide my path, including, but certainly not limited to,

Kateryn Rochon, Jordan Bannerman, Terry Galloway, Kyle Bobiwash, and Rob Currie.

There are also four people without whose friendship and mentoring I would not be where

I am today; Marianne Hardy, Joy Stacey, Bob Wrigley, and the late Nancy Loadman.

I must also acknowledge the help and academic support I received from my committee

members, Anne Worley, and Richard Westwood. I am also indebted to Richard for part of the

idea of this project, as well as for helping spark my passion for insects. I also thank Bethanne

Bruninga-Socolar for assistance.

Several people assisted with identification of my specimens, including Jason, Amber

Bass and Joel Gardner for my bees, and Adam Brunke, Stewart Peck, Don Chandler, Andrew

Smith, Serge Laplante, Anthony Davies, and Margaret Thayer for my beetles. I am very

fortunate to have your assistance, especially given the immense number of specimens and such a

tight deadline.

vi

I received funding from several sources throughout this program, and I am grateful to

each and every one, including funding from Jason Gibbs, the University of Manitoba Graduate

Fellowship, scholarships funded by Marlene Hoogmoed, and the Entomological Society of

Canada.

I must thank Nature Conservancy Canada, Nature Manitoba, The Manitoba Heritage

Corporation, and their partners for land access and research permits, especially Julie Pelc,

Melissa Grantham, and Tim Teetaert from the NCC for their assistance. I would also like to

thank the staff at the Glenlea dairy farm for collecting fresh cow dung, as well as the staff at

Freshwater Fish for providing me with fish skin.

I would like to thank my friends and family outside of school for their support during this

period of my life as well, including Larry Sereacki, Will Andrushko, Angela Reimer, Brenna

Sorokowski, as well as my mom, dad, sister, and cats, especially Earl.

Lastly, I would like to thank my partner Nabeel for being there for me through my ups

and downs and always cheering me up. This work would not be possible without your love and

friendship.

If I have forgotten people who have helped make this project, I am truly sorry. Please

know that I am nevertheless grateful for your assistance.

vii

viii

TABLE OF CONTENTS

MANAGEMENT AND LANDSCAPE EFFECTS ON BENEFICIAL TALLGRASS

PRAIRIE INSECTS – A STUDY OF BEES AND BEETLES ............................................... i

ABSTRACT.......................................................................................................................... iii

ACKNOWLEDGMENTS ..................................................................................................... v

TABLE OF CONTENTS.................................................................................................... viii

LIST OF TABLES................................................................................................................. x

LIST OF FIGURES ............................................................................................................. xii

Summary of thesis ................................................................................................................. 1

Research objectives and hypotheses ...................................................................................... 2

Chapter 1: Literature Review................................................................................................ 4

1.1 Tallgrass Prairie ................................................................................................................ 4

1.2 Prairie Response to Disturbance .......................................................................................... 6

Burning ............................................................................................................................. 6

Grazing ............................................................................................................................. 7

Mowing/haying .................................................................................................................. 7

1.3 Insect Response to Prairie Disturbance ................................................................................. 8

1.3.1 Herbivores ................................................................................................................. 8

1.3.2 Predators/Scavengers ................................................................................................... 9

1.3.3. Pollinators................................................................................................................10

Butterflies.........................................................................................................................10

Flies ................................................................................................................................11

Bees ................................................................................................................................12

1.4 Landscape Effects on Bees and Beetles................................................................................13

1.5 Bees ...............................................................................................................................14

Bee Taxonomy, Diversity and Evolutionary History................................................................14

1.6 Haplogastran Beetles ........................................................................................................16

Beetle Taxonomy, Diversity, and Evolutionary History ...........................................................16

1.7 Overview of Thesis ..........................................................................................................18

1.8 References ......................................................................................................................18

Chapter 2: Spatially diverse, unmanaged sites contain greater wild bee richness and

diversity than grazed sites in the tallgrass prairie ecosystem .............................................. 30

2.0 Abstract ..........................................................................................................................30

2.1 Introduction ....................................................................................................................30

ix

2.2 Methods .........................................................................................................................35

2.2.1 Site selection .............................................................................................................35

2.2.2 Sampling Procedure ...................................................................................................36

2.2.3 Processing Bees .........................................................................................................38

2.2.4 Ground cover characteristics ........................................................................................38

2.2.5 Landscape mapping ....................................................................................................39

2.3 Data analyses ..................................................................................................................39

2.3.1 Bee diversity & community composition .......................................................................39

2.3.2 Modelling .................................................................................................................41

2.4 Results ...........................................................................................................................42

2.4.1. Bee captures .............................................................................................................42

2.4.2 Community Composition - Distance-based Redundancy Analysis & perMANOVA .............42

2.4.3. Abundance, Species Richness, and Hill’s Shannon Diversity Modeling .............................44

2.4.3.1. Effect of Disturbance Type ......................................................................................45

2.4.3.2. Effect of landscape characteristics .............................................................................50

2.4.3.3. Effect of groundcover characteristics .........................................................................50

2.4.4 Rarefaction ...............................................................................................................50

2.5 Discussion ......................................................................................................................54

2.6 Conclusion ......................................................................................................................60

2.7 References ......................................................................................................................60

2.8 Chapter 2 Supplementary Material ......................................................................................72

Bee Model Tables ..............................................................................................................74

Chapter 3: Relationship between the two data chapters (Chapters 2 and 4) ...................... 84

Chapter 4: Fire and grazing present trade-offs for an ancestrally detritivorous group of

beetles (Coleoptera: Haplogastra) in tallgrass prairie......................................................... 85

4.0 Abstract ..........................................................................................................................85

4.1 Introduction ....................................................................................................................85

4.2 Methods .........................................................................................................................89

4.2.1 Site selection .............................................................................................................89

4.2.2 Landscape Mapping ...................................................................................................91

4.2.3 Identification/Data analysis .........................................................................................92

4.3 Results ...........................................................................................................................94

4.3.1 Community Composition - Distance-based Redundancy Analysis/perMANOVA.................95

4.3.2 Models....................................................................................................................... 100

4.3.2.1 Disturbance effects ................................................................................................ 100

x

4.3.2.2 Landscape effects .................................................................................................. 103

4.3.2.3 Ground cover effects .............................................................................................. 104

4.3.3 Rarefaction Curves ................................................................................................... 104

4.3.4 Indicator Species Analysis......................................................................................... 107

4.4 Discussion .................................................................................................................... 108

4.5 Conclusion .................................................................................................................... 115

4.6 References .................................................................................................................... 116

4.7 Chapter 4 Supplementary Materials .................................................................................. 127

Haplogastran Beetles Model Tables .................................................................................... 129

Chapter 5: Conflicting response of pollinator, decomposer, and predator insects to

disturbance and landscape in northern tallgrass prairie................................................... 139

5.0 Abstract ........................................................................................................................ 139

5.1 Introduction .................................................................................................................. 140

5.2 Disturbance Type Effects ................................................................................................ 140

5.3 Landscape Composition Effects........................................................................................ 143

5.4 Ground Cover Effects ..................................................................................................... 143

5.5 New and Interesting Records ........................................................................................... 144

5.6 Future Directions ........................................................................................................... 144

5.7 Conclusion .................................................................................................................... 145

5.8 References .................................................................................................................... 145

LIST OF TABLES

Table 2.1 Summary of perMANOVAs performed on the Bray-Curtis distance of different bee

communities ………………………………………………………………………………….. 44

Table 2.2 Results of GLMMs performed on bee bowl capture abundance, species richness, and

Hill’s Shannon Effective Species Number …………………………………………………... 74

Table 2.3 Results of GLMMs performed on ground-nesting bee bowl capture abundance, species

richness, and Hill’s Shannon Effective Species Number ………………….………………… 75

Table 2.4 Results of GLMMs performed on stem-nesting bee bowl capture abundance, species

richness, and Hill’s Shannon Effective Species Number …………………………………… 76

Table 2.5 Results of GLMMs performed on blue vane bee capture abundance, species richness,

and Hill’s Shannon Effective Species Number ……………………………………………... 77

Table 2.6 List of species caught in bee bowls in 2017 ……………..………………………. 79

xi

Table 2.7 List of species caught in bee bowls in 2018 ………………………………………. 80

Table 2.8 List of species caught in blue vane traps in 2018 …………………………………. 82

Table 2.9 List of study sites and coordinates ………………………………………………….83

Table 4.1 Summary of amount of variation explained by constrained (Disturbance type) factors

in dbRDA for the total beetle communities in both years …………………………………… 95

Table 4.2 Summary of the effect of ‘Disturbance Type’ on beetle dissimilarities in distance-

based redundancy analyses ………………………………………………………………….. 96

Table 4.3 Summary of perMANOVAs performed on the Bray-Curtis dissimilarity of different

beetle communities in 2017 & 2018 ……………………………………………………….. 98

Table 4.4 Results of Indicator Species Analysis on beetle species by disturbance type …… 107

Table 4.5 Results of GLMMs performed on haplogastran beetle pitfall trap captures …….. 129

Table 4.6 Results of GLMMs performed on predatory haplogastran beetle pitfall trap captures

………………………………………………………………………………………………. 130

Table 4.7 Results of GLMMs performed on decomposer haplogastran beetle pitfall trap captures

………………………………………………………………………………………..…….. 131

Table 4.8 Results of a GLMM performed on aleocharine beetle abundance ……..………. 132

Table 4.9 Results of Indicator Species Analysis on beetle species by site using the Indval.g

method ……………………………………………………………………………………... 133

Table 4.10 Results of Indicator Species Analysis on beetle species by site using the r.g. method

……………………………………………………………………………………………… 133

Table 4.11 List of species caught in pitfall traps in 2017 …………………………………. 134

Table 4.12 List of species caught in pitfall traps in 2018 …………………………………. 135

Table 4.13 List of study sites and coordinates ………………………………………………138

xii

LIST OF FIGURES

Figure 1.1 Map showing the distribution of the Northern Tall Grasslands ecoregion in North

America ……………………………………………………………………………………….. 5

Figure 1.2 Map showing the location of the Manitoba Tall Grass Prairie Preserve (MTGPP)

………………………………………………………………………………………………… 6

Figure 2.1 Location of the study sites showing the location of the Manitoba Tallgrass Prairie

Preserve ………………………………………………………………………………………. 36

Figure 2.2 Example of the ground level and elevated bee bowl station and blue vane

trap…………………………………………………………………………………………….. 37

Figure 2.3 Ordination plots of the Bray-Curtis distance-based redundancy analysis of bee bowl

community data ………………………………………………………………………………. 43

Figure 2.4 Ordination plot of the Bray-Curtis distance-based redundancy analysis of blue vane

trap captures …………………………………………………………………………………. 44

Figure 2.5 Effect plot showing the predicted effect of disturbance type on bee bowl trap

abundance, species richness, and Hill’s Shannon effective species number ………………… 46

Figure 2.6 Effect plot showing the predicted effect of disturbance type on ground nesting bee

species richness ………………………………………………………………………………. 47

Figure 2.7 Effect plot showing the predicted effect of disturbance type on stem nesting bee

abundance ……………………………………………………………………………………. 49

Figure 2.8 Effect plot showing the predicted effect of disturbance type on stem nesting bee

species richness ………………………………………………………………………………. 49

Figure 2.9 Rarefied curves of the entire bee bowl captured community, and the blue vane

captured community …………………………………………………………………………. 52

Figure 2.10 Rarefied curves of the ground nesting bee bowl captured community, and the stem

nesting bee bowl community ………………………………………………………………… 53

Figure 2.11 Graph showing the first two axes of a PCA performed on groundcover data …. 72

Figure 2.12 Graph of the first two axes of a principal coordinate analysis performed on percent

cover on the various landscapes found at 1 km from the centre of study sites ……………… 73

Figure 4.1 Map showing the distribution of the northern tall grasslands ecoregion in North

America ……………………………………………………………………………………… 86

Figure 4.2 Map of the distribution of the study sites relative to the town of Stuartburn …… 90

xiii

Figure 4.3 Distance-based redundancy ordination of all 2017 haplogastran beetles, predatory

beetles, and decomposer beetles …………………………………………………………….. 97

Figure 4.4 Distance-based redundancy ordination of all 2018 haplogastran beetles, predatory

beetles, and decomposer beetles ……………………………………………………………. 99

Figure 4.5 Effect plots of the predicted effect of disturbance on haplogastran beetle abundance

Hill’s Shannon effective species number …………………………………………………… 101

Figure 4.6 Effect plots of the predicted effect of disturbance type on predatory haplogastran

beetle, and decomposer beetle abundance ………………………………………………….. 102

Figure 4.7 Effect plot of the predicted effect of disturbance type on aleocharine beetle

abundance …………………………………………………………………………………... 103

Figure 4.8 Rarefied diversity curves of haplogastran beetle captures (except the subfamily

Aleocharinae) ………………………………………………………………………………. 105

Figure 4.9 Graph of the first two axes of a principal coordinate analysis performed on percent

cover on the various landscapes found at 500 m from the centre of study sites …………… 127

Figure 4.10 Graph showing the first two axes of a PCA performed on groundcover data …. 128

Figure 5.1 Bar graph representing the effect of the three disturbance types on bee bowl caught

bee abundance, species richness, and Hill’s Shannon effective species number ……………. 141

Figure 5.2 Bar graphs representing the effect of the three disturbance types on beetle abundance

in 2017 and 2018 …………………………………………………………………………….. 142

Figure 5.3 Bar graphs representing the effect of the three disturbance types on beetle species

richness in 2017 and 2018, and on Hill’s Shannon effective species numbers in 2017 and 2018

………………………………………………………………………………………………... 142

1

Summary of thesis

The focus of this thesis is primarily to understand how beneficial insect guilds respond to

common disturbance types in the endangered tallgrass prairie ecosystem. The thesis is organized

such that the first chapter presents a literature review concerning the current knowledge of how

disturbance/management techniques and landscape context affect prairie vegetation and insect

communities. The second chapter is in a manuscript format and is chiefly concerned with an

experiment I conducted that involved catching wild bees in sites that differed based on the

disturbance type that they were recently exposed to; with some sites having experienced a recent

burn, some sites being recently grazed, and other sites not experiencing a recent disturbance. I

also measured landscape and groundcover variables to understand what factors are influencing

the abundance, richness, and diversity of wild bees, and their differing nesting guilds in the

tallgrass prairie. The third chapter is a short explanation on how the two data manuscripts

(chapters 2 and 4) relate and compliment one another. Similar to the second, the fourth chapter

deals with haplogastran beetles, specifically looking at differences in response between predatory

and decomposer beetle guilds. The fifth chapter represents a summary of the previous chapters

and reports on general trends found in the proceeding experiments, as well as new provincial

beetle records.

2

Research objectives and hypotheses

The primary goals of both research chapters (Chapters 2 and 4) are to understand the

combined and specific effects that prairie disturbance, regional landscape, and local scale

landcover characteristics have on two groups of beneficial insects. The second chapter examines

the background, methods, and results of an experiment designed to test these effects on wild

bees, which are one of the main groups of pollinators in the tallgrass prairie. The fourth chapter

explores the same questions aimed at an ancestrally detritivorous group of beetles that act as

either decomposers or predators. Furthermore, I wanted to understand if different guilds of wild

bees (ground nesters vs. stem nesters) and different guilds of beetles (decomposers vs. predators)

respond differently to disturbance type, landscape context, and local landcover effects.

I hypothesized that prairie sites that were exposed to fire within the last six years would

have more exposed ground and remnant wood/stems for wild bees to utilize. As well, since most

sites had gone several years since a fire, I suspected that any short-term negative effects of

burning on stem nesting bees would have since been reversed. I also suspected that increasing

amounts of forest surrounding my sites would benefit the abundance and diversity of stem

nesting bees. As bees require floral and nesting resources to survive and propagate, I suspected

that increasing amounts of blooming flowers, forbs, and bare ground around my sites would

increase bee diversity measures.

I hypothesized that beetles would be more abundant and diverse on recently grazed sites

due to suspected greater amounts of available detritus. As well, I predicted that like previous

studies, sites with greater surrounding landscape diversity would support a lower diversity of

decomposer beetles. I expected that predatory beetles would respond positively to increasing

grass cover within my sites, as such patterns have previously been established in grassland

systems.

I hope that these studies may be used by prairie land managers seeking to understand

patterns in disturbance and landscape context relating to beneficial insect guilds, which are

becoming increasingly recognized as invaluable components of prairie diversity and processes.

3

4

Chapter 1: Literature Review

1.1 Tallgrass Prairie

The North American prairies were once a vast landscape that covered the interior of the

continent. The prairie ecozone exists as a gradient, but for convenience it is divided into three

grassland regions in Canada. Shortgrass prairie occurs east of the Rockies, mixed grass prairie

mainly in Saskatchewan and western Manitoba, and tallgrass prairie (TGP) from Manitoba to

Ontario, south to Oklahoma (Sveinson 2003). These regions differ in the amount of annual

precipitation they receive, which in turn creates differences in their vegetation communities

(Sveinson 2003). The prairies were formed and maintained by natural processes and human

activity, including large herbivores, such as the American bison, Bison bison (L.), and

Indigenous-set fires (Pyne 1986, Knapp et al. 1999). These disturbances prevent woody-plant

encroachment and are essential for increasing plant community diversity (Collins and Calabrese

2012). Tallgrass prairie is the most fertile prairie type (Joyce and Morgan 1989), and reaches its

northernmost extent in southeastern Manitoba (Figure 1.1). Tallgrass prairie is characterised by

plants such as big bluestem, Andropogon gerardii Vitman and Indian grass, Sorghastrum nutans

(L.) Nash, as well as rarer species such as the western prairie fringed orchid, Platanthera

praeclara Sheviak & M.L.Bowles (Joyce and Morgan 1989). Manitoba has lost over 99% of its

former tallgrass prairie range (Samson and Knopf 1994), and recent studies suggest that the

decline is continuing (Koper et al. 2010).

5

Figure 1.1 Map showing the distribution of the Northern Tall Grasslands ecoregion in North America. Distribution is based on ecoregions defined by the World Wildlife Fund and seen in

Olson et al. (2001). Map created with SimpleMappr (Shorthouse 2010).

To curb declines, the Manitoba Tall Grass Prairie Preserve (MTGPP) has amassed over

2,000 hectares of prairie in the Rural Municipality of Stuartburn (Figure 1.2). Though this

constitutes the largest remnant of tallgrass prairie habitat in Canada, it is nevertheless

fragmented. It exists in a mosaic of agricultural land and semi-natural aspen tree stands. The

MTGPP is located adjacent to Sandilands Provincial Forest to the east and is bordered by the

Roseau River on the west. There is an extensive swamp along the northern limit, and the

southernmost extent approaches the American border. The Preserve is managed by a committee

consisting of Nature Conservancy Canada and its partners. Management practices at MTGPP

includes cattle grazing, prescribed fire and limited mowing and rarely, haying (Sveinson 2003).

Mowing occurs in other prairie preserves in addition to prescribed fires and cattle and bison

grazing (Santerre 2010).

6

Figure 1.2 Map showing the location of the Manitoba Tall Grass Prairie Preserve (MTGPP) in

relation to Manitoba’s capital of Winnipeg and neighbouring states. Map was created using

SimpleMappr (Shorthouse 2010).

1.2 Prairie Response to Disturbance

Burning

Several biotic and abiotic changes to the prairie landscape accompany the different types

of management disturbance. Though burning leads to loss of aboveground biomass initially, it

ultimately results in greater plant productivity (Briggs and Knapp 1995, Wagle and Gowda

2018). Solar penetration is a main driver of productivity through increased temperature of soil

following litter removal. Increased soil temperature and decreased litter layer leads to earlier

greening compared to unburnt plots (Ehrenreich and Aikman 1963, Wagle and Gowda 2018).

Though burning can reduce the availability of soil nitrogen, prairie plants compensate for this by

increased root growth allocation in burned sites compared to controls (Johnson and Matchett

2001, Wagle and Gowda 2018). Fire can improve palatability to cattle and increase the width and

number of leaves on prairie grasses, which could affect herbivorous insects (Kansas State

College of Agricultural and Applied Science 1934, Wagle and Gowda 2018). The frequency and

timing of fires influence the ratios of C4/C3 grasses and forbs, along with plant species richness.

C3 and C4 grasses differ in their photosynthetic pathways and generally exhibit different

environmental preferences, with C4 grasses being favoured in hotter, drier environments

compared to C3 species (Pau et al. 2013). Forbs are non-woody, non-grass flowering plants that

7

constitute the bulk of the floral diversity in TGP (Gibson and Hulbert 1987). Spring burning is

associated with a loss of species richness and abundance of dicot forbs (Gibson and Hulbert

1987, Fay 2003). Winter and fall burnings are better for forbs and C3 grasses than late spring

burns (Towne and Craine 2014, Wagle and Gowda 2018). Unburned sites are more buffered

against the effects of draught due to their litter layer providing protection from soil evaporation

(Weaver and Rowland 1952, Wagle and Gowda 2018). Burning can also reduce soil fungi

(Dooley and Treseder 2012, Wagle and Gowda 2018), which could affect the arthropods that

feed on them.

Grazing

Grazing increases plant species richness in tallgrass prairie (Hickman et al. 2004, Wagle

and Gowda 2018), though admittedly these results were obtained from more southerly TGP

remnants. This has been attributed to the changes in soil nutrients associated with grazing;

namely nitrogen and carbon availability (Manning et al. 2017, Wagle and Gowda 2018). Grazing

has been shown to reduce the dominance of grasses and increase the numbers of flowering forbs

in grassland ecosystems (Gibson and Hulbert 1987, Hartnett et al. 1997, Coppedge and Shaw

1998, Fay 2003). Plant response can vary depending on the interaction between fire and grazing,

and plant species richness is maximized with a combination of the two management practices,

namely grazing with infrequent fire (Collins and Calabrese 2012, Wagle and Gowda 2018).

Some studies suggest that the greater the stocking density in TGP, the greater the beneficial

effects of increasing plant species richness and ratios of forbs/C3 grasses : C4 grasses, though

this was achieved with bison grazing (Wagle and Gowda 2018).

Mowing/haying

Mowing is known to play a large role in plant species composition in TGP, with some

studies showing that mowing leads to increased levels of exotic species (Gibson et al. 1993).

Studies into the effects of mowing on vegetation diversity in TGP have produced mixed results.

In some studies, mowed plots experienced low levels of diversity compared to plowing or

burning (Netherland 1979). As well, making the switch from annual mowing to annual burning

leads to increases in C3 and C4 grasses, forb, and woody plant density in TGP (Rooney and

Leach 2010). However, mowing led to increased biomass and increased species richness in

seeded prairie restorations (Dowhower et al. 2020), and plant productivity is increased by both

8

mowing and burning in TGP (Dickson 2019).Other studies show that late season mowing is

known to increase plant diversity in TGP (Dee et al. 2016). In fact, mowing has been shown to

increase TGP plant species richness negatively impacted by burning and fertilization (Collins et

al. 1998). While mowed sites tended to have less forb and plant cover than burned or untreated

sites in a TGP experiment, mowing eliminated woody plants as effectively as fire, and overall

plant diversity was not diminished (Van Dyke et al. 2004). Additionally, annual clipping has

been shown to be as effective as annual burning at promoting native warm season grasses over

exotic cool season grasses in TGP (Smart et al. 2013).

1.3 Insect Response to Prairie Disturbance

Insects are the dominant form of animal life in terrestrial ecosystems, thus, naturally

many ecological processes are insect mediated. Of these processes, those that directly benefit

humanity have been dubbed ecosystem services (Harrington et al. 2010). Various insect groups

have been shown to respond to the disturbances that form and maintain the tallgrass prairie,

largely, but not exclusively linked to their feeding guild.

1.3.1 Herbivores

Herbivores can strongly influence plant communities, and they are a large component of

the insect biodiversity in prairies. Herbivore response to management depends on the taxon and

type of disturbance. Cicada nymphs in the soil respond differently to burns by genus, with

Cicadetta preferring burned plots, whereas Tibicen heavily favoured unburned plots (Callaham et

al. 2003). Free-living hemipterans in an Illinois prairie show the greatest species richness on

unburned sites, which led to recommendations that burns should be applied only once every 3 –5

years (Wallner et al. 2012). Grasshoppers have been the subject of numerous studies in the

prairies, due to their potential economic importance. Evans (1984) found that grasshopper

diversity peaks at sites with intermediate fire frequencies i.e. spring burning every 2 or 4 years.

In mixed grass prairie, Branson and Vermeire (2016) found that late summer fire leads to a 36 –

53% decrease in grasshopper density, however responses are species dependent. Forb -feeding

and mixed-feeding grasshopper assemblages increase under infrequent fire intervals, and overall

species richness of grasshoppers is greatest at infrequent fire frequencies relative to annual or

biennial burning (Evans 1988). In contrast, Joern (2005) found that fire frequency does not affect

grasshopper species richness, but rather it is grazing that exerts a significant influence.

9

Bison grazing has been shown to increase the abundance and richness of tallgrass prairie

herbivorous arthropods (Moran 2014). Sites grazed by bison show significantly higher species

richness and Shannon diversity of grasshoppers compared to ungrazed sites (Joern 2005).

Ungulate grazing provides grasshoppers a form of resilience to the loss of host species by

increasing the complexity of grasshopper-plant networks (Welti et al. 2019). Herbivorous beetle

larvae are more prevalent when nutrients associated with animal dung (nitrogen and phosphorus)

are added to the soil (Callaham et al. 2003).

Mowing resulted in lower density of herbivorous beetle larvae and reduced the

abundance of Cicadetta spp. nymphs in TGP soil (Callaham Jr et al. 2002, 2003). As well,

mowing has been shown to reduce other soil dwelling hemipteran densities (Seastedt and Reddy

1991). Herbivorous and carnivorous insect densities have been shown to decline with increasing

mowing intensity in TGP, though interestingly, saprophagous insect densities were not affected

(Todd et al. 1992).

1.3.2 Predators/Scavengers

Ground beetles (Carabidae) have been the focus of several studies aimed at understanding

how insect communities respond to disturbance. Barber et al. (2017) concluded that prescribed

burning negatively affects large species of ground beetles. Cook and Holt (2006) demonstrated

that fire frequency has negligible effects on overall ground beetle diversity, but does alter the

relative frequency of common species. Ground beetle abundance and species richness peaks in

the year after a spring burn, with abundance diminishing year after year (Larsen and Work

2003). However, when compared to remnant prairie, diversity is highest after many years post-

fire (Larsen and Work 2003). Similarly, research conducted in a European grassland found a

three-fold increase in ground beetles in plots that had been burned, though the composition was

simpler overall and consisted of a high number of pioneer species (Samu et al. 2010). A

Manitoba study by Roughley et al. (2010) found that ground beetle species responded

individually to fire treatments and overall community composition was not affected in the long

term. They also found that it took the ground beetle community four years to completely recover

and suggest that using fire as a conservation tool should be deployed at different scales and

seasons to maximize species diversity. An experiment on an oats planting found that burning

negatively affected species diversity of herbivorous beetles, but had no effect on carnivorous

10

beetles (Bulan and Barrett 1971). Loss of litter has been shown to reduce the abundance of

detritivores in the months following a fire (Seastedt 1984, Seastedt et al. 1986, Kral et al. 2017).

However other studies have found that reduced litter led to greater numbers of largely predatory

beetles (Tester and Marshall 1961). Rove beetles have been found to be more numerous on

burned sites in TGP (Van Amburg et al. 1981, Warren et al. 1987), which is consistent with the

finding that generalist predators respond positively, and are drawn to sites that have recently

experienced a fire (Kral et al. 2017).

Several studies have been conducted on grassland ants (Hymenoptera: Formicidae) and

their response to disturbance. Moranz et al. (2013) looked at ant functional groups and found that

generalist ant species were more abundant on sites subjected to both grazing and burning, than to

burning alone. They also found that one ant species achieved competitive dominance in sites that

had been burned. Another study found that neither ant community composition nor species

richness could be characterized by the amount of time since last fire in the prairie ecosystem

(Menke et al. 2015).

Moran (2014) found that arthropod carnivores were more abundant and speciose under

bison grazing. In a study of macroinvertebrates, beetle abundance (including predators,

scavengers, and herbivores) was higher in grazed plots than in plots that experienced both fire

and grazing, while hymenopteran abundance (which included predators, parasitoids, and

herbivores) was similar between the two (Doxon et al. 2011). Leiodids, or round fungus beetles

are predicted to be more abundant at sites with greater litter depth because of the abundance of

food resources present (Chandler and Peck 1992). Beetles associated with decomposition rely on

the availability of the scarce resource of decomposing organic matter to feed and reproduce. The

focus of grazing regime studies has been largely focused on herbivores and pollinators, and thus

there are obvious gaps in our understanding on if and how grazing affects predatory and

saprophilic insects in the tall grass prairie.

1.3.3. Pollinators

Butterflies

11

Butterflies are the most well-studied pollinator in tallgrass prairies, although their role is

perhaps better characterised as herbivores. There is a lack of consistency in prairie butterfly

disturbance findings. Moranz et al. (2014) report that prescribed burning is compatible with

conservation efforts, while grazing reduced nectar sources and regal frittilary, Speyeria idalia

(Drury), density. Butterfly abundance was lowest on sites that were subjected only to a burn and

highest in sites that experienced both fire and grazing, though species diversity was highest on

burned sites (Vogel et al. 2007). For butterflies, diversity and type of disturbance is key. Grazing

affected the abundance of individual species and likely added to the broadscale diversity of the

butterfly community (Bendel et al. 2018). Prescribed fires every 3–5 years increases habitat

quality and abundance of regal fritillaries (Henderson et al. 2018). Furthermore, fire can generate

an emergent population of milkweed for migrating monarch butterflies, Danaus plexippus (L.)

(Baum and Sharber 2012). Man-made disturbances can lead to generalist species becoming more

abundant than specialists (Swengel et al. 2011). For this reason, consistency and diversity of

disturbance types are necessary for conserving specialist butterflies in tallgrass prairie habitats

(Swengel 1998). Generalist butterflies have been shown to respond best to frequent and intense

management, while random, infrequent wildfires are more useful than rotational fires for

specialists (Swengel 1998). Sites without fire disturbance are important as they house a greater

abundance of specialist butterfly species (Swengel and Swengel 2007).

Mowing, at least at infrequent intervals has been shown to positively affect prairie

butterflies. Infrequent mowing resulted in increased abundance, richness, and Shannon diversity

of grassland butterflies than continuously grazed sites in the southern Appalachians (Smith and

Cherry 2014). Mowing has also been shown to positively affect monarch butterflies, with

mowing twice leading to greater egg numbers being laid on the host milkweed (Fischer et al.

2015). While mowing is known to be preferable to grazing and burning in conserving specialist

prairie butterflies (Swengel 1996, 1998), it can nonetheless lead to reductions in extremely

endangered specialists such as the Dakota Skipper, Hesperia dacotae (Skinner) (McCabe 1981).

Overall, it is recommended that large mows be avoided in favour of localized smaller cuts

(Swengel 2001).

Flies

12

Pollinating flies are an important component of TGP, accounting for 60% of all floral

visits in Manitoba TGP (Robson 2008). In fact, differences in climate and pollinator preferences

has led to floral morphological and community differences that help distinguish the largely bee-

pollinated drier fescue and mixed grass prairie, and the wetter, largely syrphid pollinated TGP

(Robson et al. 2019). Unfortunately, flies are often neglected in grassland insect studies because

of their difficult taxonomy and incredible abundance (Paiero et al. 2010). In more southerly

prairie, syrphids (Diptera: Syrphidae) appear to be more important pollinators to annual flowers

than perennials, but still account for 9% of all floral visits in an Illinois prairie (Parrish and

Bazzaz 1979). Studies examining the role of prairie management on flies have produced mixed

results (Panzer 2002). Some studies have shown negative, or slightly negative effects of fire on

fly abundance (Rice 1932, Anderson et al. 1989). Others have shown that fire has positive effects

on abundance (Nagel 1973) and species richness (Hartley et al. 2007) of prairie flies. Globally,

pollinating fly abundance and richness tends to respond positively to intermittent fires, especially

to wildfires (Carbone et al. 2019). In TGP, flies that develop in the soil were more abundant after

a summer burn than a spring burn (Kirkwood et al. 2000). Interestingly, experimental

suppression of mycorrhizal symbiotes of grassland plants created a shift of the pollinator

community via differences in floral morphology from large bodied bees to smaller bees and flies

(Cahill Jr. et al. 2008). Under grazing, flies may be of greater importance as pollinators due to

the presence of cow dung serving as larval substrate (Vanbergen et al. 2014).

Bees

Fire may cause direct mortality on nesting bees. Nests greater than 10 cm below ground

should be safe from fire (Cane and Neff 2011), but shallow nests and stem nesters are less

protected. However, burning sites has been shown to prolong the flowering season, which may

balance out the initial high mortality (Wrobleski and Kauffman 2003, Mola and Williams 2018).

A few studies have found no difference in bee abundance between grazing and burning

disturbances, however studies comparing disturbance treatments to undisturbed reference sites

are lacking (Smith et al. 2016, Pennarola 2019). There is some indication that stocking rate of

cattle may affect bee nutrition via potential changes in quantity or quality of floral resources

(Smith et al. 2016). There is a lot of overlap in dietary requirements amongst native bees and

grazing ungulates such as cattle (Debano et al. 2016). In Missouri, grazing and haying had

detrimental effects on both the bee community and the soil and floral resources required by bees

13

(Buckles and Harmon-Threatt 2019). Kimoto et al. (2012) showed grazing intensity did not

affect bee abundance, although it did affect species composition. A recent study on bees in

tallgrass prairie found that neither fire nor grazing affected composition of bee communities

(Pennarola 2019). Grazed habitats offer pollinators a greater number of flower species, while

simultaneously increasing the size and diversity of floral visitation networks (Vanbergen et al.

2014). However, networks have less nestedness in grazed habitat, and thus they are more prone

to loss of specialist pollinator and plant species (Vanbergen et al. 2014). Though not directly tied

to the management of tallgrass prairie, introduced, stocked bees, such as the western honey bee

(Apis mellifera L.) may negatively impact wild bees via resource competition, pathogen

transmission, and change in plant community competition (Mallinger et al. 2017). As the

MTGPP exists in an agricultural mosaic, it is not unreasonable to assume that honey bees may be

affecting the wild prairie bees in one or more ways.

1.4 Landscape Effects on Bees and Beetles

Areas dominated by simplified, conventional agricultural landscapes have been shown to

negatively impact wild bee abundance and diversity (Kennedy et al. 2013, Connelly et al. 2015).

Urban and agricultural habitats support fewer rare species, which can have a large effect on

overall diversity (Harrison et al. 2017). The diversity and composition of native bee communities

vary according to proximity to forested areas in tallgrass prairie systems, with rarified species

richness increasing with increasing percentage of woodlands within a 500 m radius (Griffin et al.

2017). Olynyk (2017) found that increasing unusable habitat within the landscape negatively

affected bee abundance in a prairie/agricultural landscape. Lane et al. (2020) found that

increasing levels of surrounding agricultural landscape did not negatively affect wild bee β -

diversity. Configuration at the local scale, but not at the landscape scale, was important for bee

abundance (Graham et al. 2017, Graham and Nassauer 2019). Pollinating flies and bees respond

positively to landscape heterogeneity (Denning and Foster 2018). In Europe, impervious

surfaces, such as those encountered with urbanization, negatively affect wild bee abundance and

richness (Geslin et al. 2016). In restored grasslands, bee species richness increased with forb

cover, as well as the amount of semi-natural landscapes surrounding the grassland (Crist and

Peters 2014). Bumble bee diversity is most strongly correlated with the floral availability at a

scale of 500 m, while their abundance was linked to a scale of 700 m in the tallgrass prairie

(Hines and Hendrix 2005). It will thus be useful to determine if proximity to both natural and

14

anthropogenically modified habitat, such as those found in simple agricultural landscapes, affects

wild bee diversity and abundance in the Manitoba Tallgrass Prairie Preserve. Of different insect

feeding guilds in the tall grass prairie, predators may be more susceptible to regional (patch

shape) landscape change than lower trophic levels (Stoner and Joern 2004).

Bossenbroek et al. (2004) found that beetle community composition was affected by

environmental variables at both local and landscape scales in North American grasslands;

however, they point out that the influence of the scale factors were less consistent in explaining

beetle community composition in tall grass prairie due to its comparatively more heterogenous

makeup. Predatory beetles in restored grasslands had greater species richness as grass cover

increased at the local scale, and as semi-natural habitats surrounding the grassland increased at

the landscape scale (Crist and Peters 2014). Dung beetle species richness responds negatively to

greater landscape variety (Reynolds et al. 2018).

1.5 Bees

Bee Taxonomy, Diversity and Evolutionary History

Bees first appeared in the Cretaceous Period following the evolution of angiosperms, on

which they rely for pollen and nectar (Danforth et al. 2013). Bees are believed to have arisen

from within the apoid wasp family Crabronidae (Branstetter et al. 2017, Peters et al. 2017, Zheng

et al. 2018). Bees are composed of two or three clades split either with the long-tongued bees

(Apidae + Megachilidae) separated from the short-tongued bees (Andrenidae + Colletidae +

Halictidae + Melittidae (Zheng et al. 2018) or with Melittidae being sister to the other two

groups (Danforth et al. 2013, Branstetter et al. 2017, Peters et al. 2017). The sister group to the

Colletidae are the Stenotritidae (Branstetter et al. 2017), which is the smallest family of bees and

is restricted to Australia.

Bees vary greatly in their nesting, provisioning, and social structures (Michener 1974).

Most bees are solitary ground-nesters who provision their larvae with pollen and nectar from a

wide variety of flowering plants (Danforth et al. 2019). Some are pollen specialists dependent on

one (monolectic) or few (oligolectic) species of plants (Cane and Sipes 2006, Fowler 2016).

Some solitary nests are found in aggregations. “Communal” nests have multiple females, but

each provision only for their own offspring. Reproductive division of labour such that one

female, the queen, dominates egg laying is characteristic of semisocial and eusocial nests. Nests

15

are “semisocial”, when the worker caste are sisters of the queen (Michener 1974). Matrilineal

nests with a queen and daughter worker(s) are termed “eusocial”. “Primitively” eusocial nests

have queens (gynes) that could theoretically survive on her own, and “highly” eusocial nests

have a queen that cannot survive unassisted and requires workers to help found a new colony

(Michener 1974, 2007). Some species appear to produce a specialized male “soldier” caste with

large heads and short wings to guard the nest, however this has only been documented in a few

species (Michener 1974, Danforth et al. 2019). Many species do not nest in the ground, but

instead nest in natural cavities such as those found in hollowed out stems or reeds. Brood

parasitism has evolved in many bee lineages and can include species that lay eggs in solitary bee

nests (cleptoparasites), or social parasites who lay their eggs and take over social colonies

(Michener 2007).

Manitoba is home to six of the seven families that comprise the extant bees. Only one

species of melittid is recorded from Manitoba, Macropis nuda (Provancher). Unlike most bees,

Macropis spp. specialize on collecting floral oil in addition to pollen from yellow loosestrife ,

Lysimachia spp., to provision their larvae (Rozen and Jacobson 1980).

The family Apidae includes the honey bee, Apis mellifera, and the bumble bees, Bombus.

Apidae is the largest bee family, with a remarkable diversity including the small carpenter bees,

Ceratina spp., long-horned bees of the recently erected subfamily Eucerinae (Bossert et al.

2019), and a myriad of cuckoo bees found in the subfamily Nomadinae. The family

Megachilidae includes the leaf-cutter bees, Megachile spp., which derive their common name for

the habit of some species lining their nests with sections of leaves that they have chewed off. The

family also includes the mostly spring flying Osmia spp., resin bees (Dianthidium spp.), as well

as two genera in Manitoba that are strictly cleptoparasitic, Stelis and Coelioxys.

Of the short-tongued bees, some of the most abundant belong to the family Halictidae,

which are known as sweat bees because of their habit of landing on skin and lapping up salt-rich

excretions. Members of the nominate subfamily are mostly pollen generalists (Wcislo and Cane

1996), such as most Lasioglossum and Halictus, however some halictids are specialists, such as

Dufourea spp. (Dumesh and Sheffield 2012). Members of one genus in Canada, Sphecodes, or

blood bees, are exclusively cleptoparasitic (Michener 1978). Andrenidae includes many spring-

flying bees and many pollen specialists (Danforth et al. 2019). These bees most often nest

16

solitarily, or in aggregations in the soil (Wilson and Messinger Carril 2016). The family

Colletidae is represented by two genera in Manitoba that both line their nest with a naturally

produced cellophane-like material (Wilson and Messinger Carril 2016), Colletes spp. are usually

honey bee sized, and are superficially similar to Andrena spp. The other genus, Hylaeus, are

quite distinct from Colletes and are known as masked bees because of the yellow markings on

their face. Hylaeus have a unique way of transporting the pollen they collect, storing it in an

internal crop as opposed to carrying it on scopa like most other bees (Michener 2007).

1.6 Haplogastran Beetles

Beetle Taxonomy, Diversity, and Evolutionary History

Haplogastra is a proposed clade that includes the clades Staphyliniformia and

Scarabaeiformia. The Haplogastra are thought to have evolved as detritus feeders, which is a

habit and lifestyle retained in most of its lineages (McKenna et al. 2015). This clade is supported

by molecular and morphological evidence (Kolbe 1908, McKenna et al. 2015) (but see Beutel

and Leschen 2005). Staphyliniformia includes the rove beetles (Staphylinidae), the largest family

of beetles (and perhaps all living things) in the world, as well as the carrion beetles (Silphidae),

round fungus beetles (Leiodidae), feather-winged beetles (Ptiliidae), clown or hister beetles

(Histeridae), and water scavenger beetles (Hydrophilidae). The Scarabaeiformia includes scarabs

(Scarabaeidae), stag beetles (Lucanidae), bess beetles (Passalidae), and hide beetles (Trogidae)

and several other families.

The rove beetles (Staphylinidae) are so diverse it is difficult to describe them as a whole,

however generally the majority are predacious or saprophagous (Bohac 1999). The most visible

of the rove beetles, and those most readily sampled, are those that are quick running and live

above the soil surface (Irmler and Lipkow 2018). The rove beetles are now believed to be

paraphyletic in regards to the carrion beetles (Silphidae), with molecular evidence placing

silphids well within the Staphylinidae (McKenna et al. 2015). Carrion beetles are divided into

two subfamilies, the Silphinae, and the Nicrophorinae. The Silphinae are adapted for feeding on

large vertebrate carcasses and developing in the soil nearby (Anderson and Peck 1985). The

Nicrophorinae are known as the burying, or sexton beetles, and have shorter elytra and are

generally more streamlined. The majority are adapted to finding small vertebrate carcasses and

as a team dragging and burying it under the surface, where they remove the fur or feathers and

17

lay their eggs to develop inside the carrion (Anderson and Peck 1985). Interestingly, the

nicrophorines engage in parental care, tending and feeding the developing larvae, a trait not often

seen in insects, dubbed ‘subsocial’ behaviour (Costa 2006). Other members of the

Staphylinoidea include the feather-winged beetles (Ptiliidae) and the round fungus beetles

(Leiodidae). The feather-winged beetles include the smallest of all beetles. Their winged forms

are characterised by unique plumose wings not seen in other beetles. Most ptiliids are believed to

feed on the microflora present on rotting organic matter (Dybas 2014). Round fungus beetles

include some of the most specialized of all beetles, including the ‘Beaver Parasite Beetle’,

Platypsyllus castoris Ritsema, that lives and feeds on skin excretions on beavers. Many others

are associated with mammal nests (Krinsky 2019). In general, leiodids feed on fungal hyphae,

which are often present on rotting substrate (Grimaldi and Engel 2005).

The Hydrophiloidea includes the water scavenger beetles (Hydrophilidae). The clown

beetles (Histeridae) and closely related families belong to the Hydrophiloidea but have

sometimes treated separately as Histeroidea. The water scavenger beetles are largely aquatic,

however the subfamily Sphaeridinae is mostly terrestrial and includes forms that regularly occur

in cow pats and other rotting material (Short and Fikacek 2013). The clown or hister beetles

(Histeridae) are rather unique looking predatory beetles that occur in such substrates as dung,

carrion, and rotting cacti, while others live as inquilines in mammal nests or ant colonies, or

under bark (Bousquet and Laplante 2006). The Scarabaeiformia are believed to have descended

from within the Hydrophiloidea, or as sister group to them (McKenna et al. 2015).

The Scarabaeoidea includes the scarabs and their close relatives. The scarabs

(Scarabaeidae) are divided into several subfamilies, with some being largely herbivorous, while

two are dung feeders (Scarabaeinae and Aphodiinae). The Scarabaeinae includes such familiar

beetles as the sacred scarab, Scarabaeus sacer L., worshipped in ancient Egypt, as well as

several local species often found in and around dung. The Scarabaeinae tend to be “rollers”, or

“tunnelers”, with some species rolling pieces of dung away to a separate location, while others

create tunnels under the dung to provision their larvae (Helgesen and Post 1967, Halffter and

Edmonds 1982). The Aphodiinae includes the genus Aphodius, which has recently been split into

several different genera, as well as several other related groups. Aphodiine dung beetles are

generally smaller and more elongate than the Scarabaeinae. They tend to be “dwellers” that

18

complete their entire life cycle within a cow pat or rich humus source (Helgesen and Post 1967,

Halffter and Edmonds 1982). The Scarabaeoidea also includes other dung feeding groups (some

Geotrupidae), as well as some that develop in rotting wood (Passalidae, Lucanidae), and one

group that specializes on the keratin left over after animal decomposition (Trogidae).

1.7 Overview of Thesis

Both bees and haplogastran beetles have been suggested as indicator species of the

quality and characteristics of the landscape due to their sensitivity to disturbance and ties to

specific landscapes (Klein 1989, Schindler et al. 2013, Dorneles Audino et al. 2014,

Klimaszewski et al. 2018, Vieira et al. 2018). The following chapters provide an overview and

discussion on the findings of an experiment designed to test the response of bee and beetle

communities to disturbance type and landscape characteristics in the tallgrass prairie. Chapter 2

will be in a manuscript format and will focus on the response of bees to prairie disturbance,

environmental variables, and landscape characteristics. Chapter 3 will be in a manuscript format

and focus on the same characteristics applied to haplogastran beetles. Finally, chapter 4 will tie

together the findings of chapters 2 and 3 and discuss implications for prairie management going

forward.

1.8 References

Van Amburg, G.L., Swaby, J.A., and Pemble, R.H. 1981. Response of arthropods to a spring

burn of a tallgrass prairie in northwestern Minnesota. Ohio Biological Survey Biology Notes, 15: 240–243.

Anderson, R.C., Leahy, T., and Dhillion, S.S. 1989. Numbers and biomass of selected insect groups on burned and unburned sand prairie. The American Midland Naturalist, 122: 151–

162.

Anderson, R.S., and Peck, S.B. 1985. The insects and arachnids of Canada part 13: the carrion beetles of Canada and Alaska (Coleoptera: Silphidae and Agyrtidae). Canadian Government Publishing Centre, Ottawa. p.

Barber, N.A., Lamagdeleine-Dent, K.A., Willand, J.E., Jones, H.P., and McCravy, K.W. 2017. Species and functional trait re-assembly of ground beetle communities in restored grasslands. Biodiversity and Conservation, 26: 3481–3498. doi: 10.1007/s10531-017-1417-6.

Baum, K.A., and Sharber, W. V. 2012. Fire creates host plant patches for monarch butterflies. Biology Letters, 8: 968–971.

Bendel, C.R., Hovick, T.J., Limb, R.F., and Harmon, J.P. 2018. Variation in grazing management practices supports diverse butterfly communities across grassland working

19

landscapes. Journal of Insect Conservation, 22: 99–111. doi: 10.1007/s10841-017-0041-9.

Beutel, R.G., and Leschen, R.A.B. 2005. Coleoptera, beetles volume 1: morphology and systematics (Archostemata, Adephaga, Myxophaga, Polyphaga partim). Handbook of

Zoology, 4: 1–567.

Bohac, J. 1999. Staphylinid beetles as bioindicators. Agriculture, Ecosystems and Environment, 74: 357–372.

Bossenbroek, J.M., Wagner, H.H., and Wiens, J.A. 2004. Taxon-dependent scaling: beetles,

birds, and vegetation at four North American grassland sites. Landscape Ecology, 20: 675–688. doi: 10.1007/s10980-004-5651-4.

Bossert, S., Murray, E.A., Almeida, E.A.B., Brady, S.G., Blaimer, B.B., and Danforth, B.N. 2019. Molecular phylogenetics and evolution combining transcriptomes and ultraconserved

elements to illuminate the phylogeny of Apidae. Molecular Phylogenetics and Evolution, 130: 121–131. doi: 10.1016/j.ympev.2018.10.012.

Bousquet, Y., and Laplante, S. 2006. The insects and arachnids of Canada part 24: Coleoptera Histeridae. NRC Research Press, Ottawa. p.

Branstetter, M.G., Danforth, B.N., Pitts, J.P., Gates, M.W., Kula, R.R., Brady, G., Branstetter, M.G., Danforth, B.N., Pitts, J.P., Faircloth, B.C., and Ward, P.S. 2017. Phylogenomic insights into the evolution of stinging wasps and the origins of ants and bees. Current Biology, 27: 1019–1025. doi: 10.1016/j.cub.2017.03.027.

Briggs, J.M., and Knapp, A.K. 1995. Interannual variability in primary production in tallgrass prairie: climate, soil moisture, topographic position, and fire as determinants of aboveground biomass. American Journal of Botany, 82: 1024–1030.

Buckles, B.J., and Harmon-Threatt, A.N. 2019. Bee diversity in tallgrass prairies affected by

management and its effects on above‐ and below‐ground resources. Journal of Applied Ecology, 56: 2443–2453. doi: 10.1111/1365-2664.13479.

Bulan, C.A., and Barrett, G.W. 1971. The effects of two acute stresses on the arthropod component of an experimental grassland ecosystem. Ecology, 52: 597–605.

Cahill Jr., J.F., Elle, E., Smith, G.R., and Shore, B.H. 2008. Disruption of a belowground mutualism alters interactions between plants and their floral visitors. Ecology, 89: 1791–1801.

Callaham Jr, M.A., Blair, J.M., Todd, T.C., Kitchen, D.J., and Whiles, M.R. 2003.

Macroinvertebrates in North American tallgrass prairie soils: effects of fire, mowing, and fertilization on density and biomass. Soil Biology & Biochemistry, 35: 1079–1093. doi: 10.1016/S0038-0717(03)00153-6.

Callaham Jr, M.A., Whiles, M.R., and Blair, J.M. 2002. Annual fire, mowing and fertilization

effects on two cicada species (Homoptera: Cicadidae) in tallgrass prairie. American Midland Naturalist, 148: 90–101.

Cane, J., and Sipes, S. 2006. Characterizing floral specialization by bees: Analytical methods and a revised lexicon for oligolecty. In Plant-Pollinator Interactions: From Specialization to

20

Generalization. Edited by N.M. Waser and J. Ollerton. The University of Chicago Press, Chicago. pp. 99–122.

Cane, J.H., and Neff, J.L. 2011. Predicted fates of ground-nesting bees in soil heated by wildfire:

thermal tolerances of life stages and a survey of nesting depths. Biological Conservation, 144: 2631–2636. doi: 10.1016/j.biocon.2011.07.019.

Carbone, L.M., Tavella, J., Pausus, J.G., and Aguilar, R. 2019. A global synthesis of fire effects on pollinators. Global Ecology and Biogeography, 28: 1487–1498. doi: 10.1111/geb.12939.

Chandler, D.S., and Peck, S.B. 1992. Diversity and seasonality of leiodid beetles (Coleoptera: Leiodidae) in an old-growth and a 40-year-old forest in New Hampshire. Environmental Entomology, 21: 1283–1293.

Collins, S.L., and Calabrese, L.B. 2012. Effects of fire, grazing and topographic variation on

vegetation structure in tallgrass prairie. Journal of Vegetation Science, 23: 563–575. doi: 10.1111/j.1654-1103.2011.01369.x.

Collins, S.L., Knapp, A.K., Briggs, J.M., Blair, J.M., and Steinauer, E.M. 1998. Modulation of diversity by grazing and mowing in native tallgrass prairie. Science, 280: 745–747.

Connelly, H., Poveda, K., and Loeb, G. 2015. Landscape simplification decreases wild bee pollination services to strawberry. Agriculture, Ecosystems and Environment, 211: 51–56. doi: 10.1016/j.agee.2015.05.004.

Cook, W.M., and Holt, R.D. 2006. Fire frequency and mosaic burning effects on a tallgrass

prairie ground beetle assemblage. Biodiversity and Conservation, 15: 2301–2323. doi: 10.1007/s10531-004-8227-3.

Coppedge, B.R., and Shaw, J.H. 1998. Bison grazing prairie patterns on seasonally burned tallgrass. Journal of Rangeland Management, 51: 258–264.

Costa, J.T. 2006. The other insect societies. The Belknap Press of Harvard University Press, Cambridge, Massachusetts. p.

Crist, T.O., and Peters, V.E. 2014. Landscape and local controls of insect biodiversity in conservation grasslands: implications for the conservation of ecosystem service providers in

agricultural environments. Land, 3: 693–718. doi: 10.3390/land3030693.

Danforth, B.N., Cardinal, S., Praz, C., Almeida, E.A.B., and Michez, D. 2013. The impact of molecular data on our understanding of bee phylogeny and evolution. Annual Review of Entomology, 58: 57–78. doi: 10.1146/annurev-ento-120811-153633.

Danforth, B.N., Minckley, R.L., and Neff, J.L. 2019. The solitary bees. Princeton University Press, Princeton. p.

Debano, S.J., Roof, S.M., Rowland, M.M., Smith, L.A., and Debano, S.J. 2016. Diet overlap of mammalian herbivores and native bees: implications for managing co- occurring grazers

and pollinators. Natural Areas Journal, 36: 458–477.

Dee, J.R., Thomas, S.M., Thompson, S.D., and Palmer, M.W. 2016. Long-term late season mowing maintains diversity in southern US tallgrass prairie invaded by Bothriochloa

21

ischaemum. Applied Vegetation Science, 19: 442–453. doi: 10.1111/avsc.12227.

Denning, K.R., and Foster, B.L. 2018. Taxon-specific associations of tallgrass prairie flower visitors with site-scale forb communities and landscape composition and configuration.

Biological Conservation, 227: 74–81. doi: 10.1016/j.biocon.2018.08.023.

Dickson, T.L. 2019. Burning and mowing similarly increase prairie plant production in the spring, but not due to increased soil temperatures. Ecosphere, 10: e02606. 10.1002/ecs2.2606. doi: 10.1002/ecs2.2606.

Dooley, S.R., and Treseder, K.K. 2012. The effect of fire on microbial biomass: a meta -analysis of field studies. Biogeochemistry, 109: 49–61. doi: 10.1007/s10533-011-9633-8.

Dorneles Audino, L., Louzada, J., and Comita, L. 2014. Dung beetles as indicators of tropical forest restoration success: Is it possible to recover species and functional diversity?

Biological Conservation, 169: 248–257. doi: 10.1016/j.biocon.2013.11.023.

Dowhower, S.L., Teague, W.R., Steigman, K., and Freiheit, R. 2020. Texas blackland prairie restoration on old-field vegetation using seeding, mowing, and burning. Arid Land Research and Management, 0: 1–17. doi: 10.1080/15324982.2020.1774941.

Doxon, E.D., Davis, C.A., Fuhlendorf, S.D., and Winter, S.L. 2011. Aboveground macroinvertebrate diversity and abundance in sand sagebrush prairie managed with the use of pyric herbivory. Rangeland Ecology & Management, 64: 394–403. doi: 10.2111/REM-D-10-00169.1.

Dumesh, S., and Sheffield, C.S. 2012. Bees of the genus Dufourea Lepeletier (Hymenoptera: Halictidae: Rophitinae) of Canada. Canadian Journal of Arthropod Identification, 20: 1–36. doi: 10.3752/cjai.2012.20.

Dybas, H.S. 2014. Featured creatures: featherwing beetles. Available from

http://entnemdept.ufl.edu/creatures/misc/beetles/featherwing_beetles.htm [accessed 21 May 2020].

Van Dyke, F., Van Kley, S.E., Page, C.E., and Van Beek, J.G. 2004. Restoration efforts for plant and bird communities in tallgrass prairies using prescribed burning and mowing.

Restoration Ecology, 12: 575–585.

Ehrenreich, J.H., and Aikman, J.M. 1963. An ecological study of the effect of certain management practices on native prairie in Iowa. Ecological Monographs, 33: 113–130.

Evans, E.W. 1984. Fire as a natural disturbance to grasshopper assemblages of tallgrass prairie.

Oikos, 43: 9–16.

Evans, E.W. 1988. Grasshopper (Insecta: Orthoptera: Acrididae) assemblages of tallgrass prairie: influences of fire frequency, topography, and vegetation. Canadian Journal of Zoology, 66: 1495–1501.

Fay, P.A. 2003. Insect diversity in two burned and grazed grasslands. Environmental Entomology, 32: 1099–1104.

Fischer, S.J., Williams, E.H., Brower, L.P., and Palmiotto, P.A. 2015. Enhancing Monarch

22

Butterfly reproduction by mowing fields of Common Milkweed. American Midland Naturalist, 173: 229–240.

Fowler, J. 2016. Specialist bees of the northeast: Host plants and habitat conservation.

Northeastern Naturalist, 23: 305–320.

Geslin, B., Le Feon, V., Folschweiller, M., Flacher, F., Carmignac, D., Motard, E., Perret, S., and Dajoz, I. 2016. The proportion of impervious surfaces at the landscape scale structures wild bee assemblages in a densely populated region. Ecology and Evolution, 6: 6599–6615. doi:

10.1002/ece3.2374.

Gibson, D.J., and Hulbert, L.C. 1987. Effects of fire, topography and year-to-year climatic variation on species composition in tallgrass prairie. Vegetatio, 72: 175–185.

Gibson, D.J., Seastedtt, T.R., and Briggs, J.M. 1993. Management practices in tallgrass prairie :

Large- and small-Scale experimental effects on species composition. Journal of Applied Ecology, 30: 247–255.

Graham, J.B., and Nassauer, J.I. 2019. Wild bee abundance in temperate agroforestry landscapes: assessing effects of alley crop composition, landscape configuration, and agroforestry area.

Agroforestry Systems, 93: 837–850. doi: 10.1007/s10457-017-0179-1.

Graham, J.B., Nassauer, J.I., Currie, W.S., Ssegane, H., and Negri, M.C. 2017. Assessing wild bees in perennial bioenergy landscapes: effects of bioenergy crop composition, landscape configuration, and bioenergy crop area. Landscape Ecology, 32: 1023–1037. doi:

10.1007/s10980-017-0506-y.

Griffin, S.R., Bruninga-Socolar, B., Kerr, M.A., Gibbs, J., and Winfree, R. 2017. Wild bee community change over a 26-year chronosequence of restored tallgrass prairie. Restoration Ecology, 25: 650–660. doi: 10.1111/rec.12481.

Grimaldi, D., and Engel, M.S. 2005. Evolution of the insects. Cambridge University Press, New York. p.

Halffter, G., and Edmonds, W.D. 1982. The nesting behavior of dung beetles (Scarabaeinae): an ecological and evolutive approach. Instituto de Ecologia, Mexico, D.F. p.

Harrington, R., Anton, C., Dawson, T.P., de Bello, F., Feld, C.K., Haslett, J.R., Kluvánkova -Oravská, T., Kontogianni, A., Lavorel, S., Luck, G.W., Rounsevell, M.D.A., Samways, M.J., Settele, J., Skourtos, M., Spangenberg, J.H., Vandewalle, M., Zobel, M., and Harrison, P.A. 2010. Ecosystem services and biodiversity conservation: concepts and a glossary.

Biodiversity and Conservation, 19: 2773–2790. doi: 10.1007/s10531-010-9834-9.

Harrison, T., Gibbs, J., and Winfree, R. 2017. Anthropogenic landscapes support fewer rare bee species. Landscape Ecology, 34: 967–978. doi: 10.1007/s10980-017-0592-x.

Hartley, M.K., Rogers, W.E., Siemann, E., and Grace, J. 2007. Responses of prairie arthropod

communities to fire and fertilizer: Balancing plant and arthropod conservation. American Midland Naturalist, 157: 92–105.

Hartnett, D.C., Steuter, A.A., and Hickman, K.R. 1997. Ecology and conservation of great plains vertebrates. In Ecological Studies 125. Edited by F. Knopf and F. Samson. Springer, New

23

York. pp. 72–101.

Helgesen, R.G., and Post, R.L. 1967. Saprophagous Scarabaeidae (Coleoptera) of North Dakota. North Dakota Insects, p.

Henderson, R.A., Meunier, J., and Holoubek, N.S. 2018. Disentangling effects of fire, habitat, and climate on an endangered prairie- specialist butterfly. Biological Conservation, 218: 41–48. doi: 10.1016/j.biocon.2017.10.034.

Hickman, K.R., Hartnett, D.C., Cochran, R.C., and Owensby, C.E. 2004. Grazing management

effects on plant species diversity in tallgrass prairie. Journal of Range Management, 57: 58–65.

Hines, H.M., and Hendrix, S.D. 2005. Bumble bee (Hymenoptera: Apidae) diversity and abundance in tallgrass prairie patches: effects of local and landscape floral resources.

Environmental Entomology, 34: 1477–1484.

Irmler, U., and Lipkow, E. 2018. Effect of environmental conditions on distribution patterns of rove beetles. In Biology of rove beetles (Staphylinidae): life history, evolution, ecology and distribution. Edited by O. Betz, U. Irmler, and J. Klimaszewski. Springer, Cham,

Switzerland. pp. 117–144.

Joern, A. 2005. Disturbance by fire frequency and bison grazing modulate grasshopper assemblages in tallgrass prairie. Ecology, 86: 861–873. doi: 10.1890/04-0135.

Johnson, L.C., and Matchett, J.R. 2001. Fire and grazing regulate belowground processes in

tallgrass prairie. Ecology, 82: 3377–3389.

Joyce, J., and Morgan, J.P. 1989. Manitoba’s tall-grass prairie conservation project. Proceedings of the North American Prairie Conferences, 34: 71–74.

Kansas State College of Agricultural and Applied Science. 1934. Effect of burning on Kansas

bluestem pastures. In Agricultural Experiment Station Technical Bulletin 88. Kansas State College of Agriculture and Applied Science, Manhattan, Kansas. p.

Kennedy, C.M., Lonsdorf, E., Neel, M.C., Williams, N.M., Taylor, H., Winfree, R., Brittain, C., Alana, L., and Cariveau, D. 2013. A global quantitative synthesis of local and landscape

effects on wild bee pollinators in agroecosystems. Ecology Letters, 16: 584–599. doi: 10.1111/ele.12082.

Kimoto, C., DeBano, S.J., Thorp, R.W., Taylor, R. V, Schmalz, H., DelCurto, T., Johnson, T., Kennedy, P.L., and Rao, S. 2012. Short-term responses of native bees to livestock and

implications for managing ecosystem services in grasslands. Ecosphere, 3: 1–19.

Kirkwood, M., Shapiro, L., and Walker, X. 2000. Edaphic prairie arthropods, with the exception of Diptera, show no reaction to seasonal burns. Tillers, 2: 47–57.

Klein, B.C. 1989. Effects of forest fragmentation on dung and carrion beetle communities in

central Amazonia. Ecology, 70: 1715–1725.

Klimaszewski, J., Brunke, A., Work, T.T., and Venier, L. 2018. Rove beetles (Coleoptera, Staphylinidae) as bioindicators of change in boreal forests and their biological control

24

services in agroecosystems: Canadian case studies. In Biology of rove beetles (Staphylinidae): life history, evolution, ecology and distribution. Edited by O. Betz, U. Irmler, and J. Klimaszewski. Springer, Cham, Switzerland. pp. 161–181.

Knapp, A.K., Blair, J.M., Briggs, J.M., Collins, S.L., Hartnett, D.C., Johnson, L.C., Towne, E.G., Knapp, A.K., John, M., Scott, L., Johnson, L.C., Towne, E.G., Hartnett, D.C., Collins, S.L., and Hartnett, D.C. 1999. The keystone role of bison in American tallgrass prairie: bison increase habitat heterogeneity and alter a broad array of plant, community, and ecosystem

processes. BioScience, 49: 39–50.

Kolbe, H.J. 1908. Mein system der coleopteren. Zeitschrift fur Wissenschaftliche Insektenbiologie, 4: 116–123.

Koper, N., Mozel, K.E., and Henderson, D.C. 2010. Recent declines in northern tall-grass

prairies and effects of patch structure on community persistence. Biological Conservation, 143: 220–229. doi: 10.1016/j.biocon.2009.10.006.

Kral, K.C., Limb, R.F., Harmon, J.P., and Hovick, T.J. 2017. Arthropods and fire: previous research shaping future conservation. Rangeland Ecology & Management, 70: 589–598.

doi: 10.1016/j.rama.2017.03.006.

Krinsky, W.L. 2019. Nest associates and ectoparasites. In Medical and veterinary entomology, Third Edit. Edited by G.R. Mullen, L.A. Durden, and J.G. King. Elsevier, London. p. 769.

Lane, I.G., Herron-sweet, C.R., Portman, Z.M., and Cariveau, D.P. 2020. Floral resource

diversity drives bee community diversity in prairie restorations along an agricultural landscape gradient. Journal of Applied Ecology, In Press: 1–30. doi: 10.1111/1365-2664.13694.

Larsen, K.J., and Work, T.W. 2003. Differences in ground beetles (Coleoptera: Carabidae) of

original and reconstructed tallgrass prairies in northeastern Iowa, USA, and impact of 3-year spring burn cycles. Journal of Insect Conservation, 7: 153–166.

Mallinger, R.E., Gaines-day, H.R., and Gratton, C. 2017. Do managed bees have negative effects on wild bees?: a systematic review of the literature. PLoS ONE, 12: 1–32.

Manning, G.C., Baer, S.G., and Blair, J.M. 2017. Effects of grazing and fire frequency on floristic quality and its relationship to indicators of soil quality in tallgrass prairie. Environmental Management, 60: 1062–1075. doi: 10.1007/s00267-017-0942-0.

McCabe, T.L. 1981. The Dakota Skipper, Hesperia dacotae (Skinner): Range and biology, with

special reference to North Dakota. Journal of the Lepidopterists’ Society, 35: 179–193.

McKenna, D.D., Farrell, B.D., Caterino, M.S., Farnum, C.W., Hawks, D.C., Maddison, D.R., Seago, A.E., Short, A.E.Z., Newton, A.F., and Thayer, M.K. 2015. Phylogeny and evolution of Staphyliniformia and Scarabaeiformia: forest litter as a stepping stone for diversification

of nonphytophagous beetles. Systematic Entomology, 40: 35–60. doi: 10.1111/syen.12093.

Menke, S.B., Gaulke, E., Hamel, A., and Vachter, N. 2015. The effects of restoration age and prescribed burns on grassland ant community structure. Environmental Entomology, 44: 1336–1347. doi: 10.1093/ee/nvv110.

25

Michener, C.D. 1974. The social behavior of the bees: a comparative study. The Belknap Press of Harvard University Press, Cambridge, Massachusetts. p.

Michener, C.D. 1978. The parasitic groups of Halictidae (Hymenoptera, Apoidea). The

University of Kansas Science Bulletin, 51: 291–339.

Michener, C.D. 2007. The Bees of the World. In Second edi. The Johns Hopkins University Press, Baltimore. p.

Mola, J.M., and Williams, N.M. 2018. Fire-induced change in floral abundance, density, and

phenology benefits bumble bee foragers. Ecosphere, 9: 1–9. doi: 10.1002/ecs2.2056.

Moran, M.D. 2014. Bison grazing increases arthropod abundance and diversity in a tallgrass prairie. Environmental Entomology, 43: 1174–1184.

Moranz, R.A., Debinski, D.M., Winkler, L., Trager, J., Mcgranahan, D.A., Engle, D.M., and

Miller, J.R. 2013. Effects of grassland management practices on ant functional groups in central North America. Journal of Insect Conservation, 17: 699–713. doi: 10.1007/s10841-013-9554-z.

Moranz, R.A., Fuhlendorf, S.D., and Engle, D.M. 2014. Making sense of a prairie butterfly

paradox: The effects of grazing, time since fire, and sampling period on regal fritillary abundance. Biological Conservation, 173: 32–41. doi: 10.1016/j.biocon.2014.03.003.

Nagel, H.G. 1973. Effect of spring prairie burning on herbivorous and non-herbivorous arthropod populations. Journal of the Kansas Entomological Society, 46: 485–496.

Netherland, L. 1979. The effect of disturbances in tall grass prairie sites on an index of diversity and equitability. The Southwestern Naturalist, 24: 267–274.

Olson, D.M., Dinerstein, E., Wikramanayake, E.D., Burgess, N.D., Powell, G.V.N., Underwood, E.C., D’Amico, J.A., Itoua, I., Strand, H.E., Morrison, J.C., Loucks, C.J., Allnutt, T.F.,

Ricketts, T.H., Kura, Y., Lamoreux, J.F., Wettengel, W.W., Hedao, P., and Kassem, K.R. 2001. Terrestrial ecoregions of the world : a new map of life on Earth. BioScience, 51: 933–938.

Olynyk, M. 2017. Effect of habitat loss, fragmentation, and alteration on wild bees and

pollination services in fragmented Manitoba grasslands. [degree] thesis, University of Manitoba, .

Paiero, S.M., Marshall, S.A., Pratt, P.D., and Buck, M. 2010. Insects of Ojibway Prairie, a southern Ontario tallgrass prairie. In Arthropods of Canadian Grasslands (Volume 1):

Ecology and Interactions in Grassland Habitats. Biological Survey of Canada, pp. 199–225. doi: 10.3752/9780968932148.ch9.

Panzer, R. 2002. Compatibility of prescribed burning with the conservation of insects in small, isolated prairie reserves. Conservation Biology, 16: 1296–1307.

Parrish, J.A.D., and Bazzaz, F.A. 1979. Difference in pollination niche relationships in early and late successional plant communities. Ecology, 60: 597–610.

Pau, S., Edwards, E.J., and Still, C.J. 2013. Improving our understanding of environmental

26

controls on the distribution of C3 and C4 grasses. Global Change Biology, 19: 184–196. doi: 10.1111/gcb.12037.

Pennarola, P. 2019. The revery alone won’t do: fire, grazing, and other drivers of bee

communities in remnant tallgrass prairie. [degree] thesis, The University of Minnesota, .

Peters, R.S., Krogmann, L., Mayer, C., Rust, J., Misof, B., Niehuis, O., Peters, R.S., Krogmann, L., Mayer, C., Donath, A., Gunkel, S., and Meusemann, K. 2017. Evolutionary history of the Hymenoptera. Current Biology, 27: 1013–1018. doi: 10.1016/j.cub.2017.01.027.

Pyne, S.J. 1986. “These conflagrated prairies” a cultural fire history of the grasslands. In The Prairie: past, present, and future: Proceedings of the ninth North American prairie conference part 6. Fire and prairie ecosystems. Edited by G.K. Clambey and R.H. Pemble.

Reynolds, C., Fletcher, R.J., Carneiro, C.M., Jennings, N., Ke, A., LaScaleia, M.C., Lukhele,

M.B., Mamba, M.L., Sibiya, M.D., Austin, J.D., Magagula, C.N., Mahlaba, T., Monadjem, A., Wisely, S.M., and McCleery, R.A. 2018. Inconsistent effects of landscape heterogeneity and land-use on animal diversity in an agricultural mosaic: a multi-scale and multi-taxon investigation. Landscape Ecology, 33: 241–255. doi: 10.1007/s10980-017-0595-7.

Rice, L.A. 1932. The effect of fire on the prairie animal communities. Ecology, 13: 392–401.

Robson, D.B. 2008. The structure of the flower–insect visitor system in tall-grass prairie. Botany, 86: 1266–1278. doi: 10.1139/B08-083.

Robson, D.B., Hamel, C., Neufeld, R., and Bleho, B.I. 2019. Habitat filtering influences plant–

pollinator interactions in prairie ecosystems. Botany, 97: 204–220.

Rooney, T.P., and Leach, M.K. 2010. Replacing hay-mowing with prescribed fire restores species diversity and conservation value in a tallgrass prairie sampled thrice : a 59 -year study. American Midland Naturalist, 164: 311–321.

Roughley, R.E., Pollock, D.A., and Wade, D.J. 2010. Tallgrass prairie, ground beetles (Coleoptera: Carabidae), and the use of fire as a biodiversity and conservation management tool. In Arthropods of Canadian Grasslands (Volume 1): Ecology and Interactions in Grassland Habitats. pp. 227–235. doi: 10.3752/9780968932148.ch10.

Rozen, J.G., and Jacobson, N.R. 1980. Biology and immature stages of Macropis nuda, including comparisons to related bees (Apoidea, Melittidae). In American Museum Novitates. The American Museum of Natural History, New York. p.

Samson, F., and Knopf, F. 1994. Prairie conservation in North America. BioScience, 44: 418–

421.

Samu, F., Kádár, F., Ónodi, G., Kertész, M., Szirányi, A., Szita, É., Fetykó, K., Neidert, D., Botos, E., and Altbäcker, V. 2010. Differential ecological responses of two generalist arthropod groups, spiders and carabid beetles (Araneae, Carabidae), to the effects of

wildfire. Community Ecology, 11: 129–139. doi: 10.1556/ComEc.11.2010.2.1.

Schindler, M., Diestelhorst, O., Härtel, S., Saure, C., Schanowski, A., and Schwenninger, H.R. 2013. Monitoring agricultural ecosystems by using wild bees as environmental indicators. BioRisk, 71: 53–71. doi: 10.3897/biorisk.8.3600.

27

Seastedt, T.R. 1984. Microarthropods of burned and unburned tallgrass prairie. Journal of the Kansas Entomological Society, 57: 468–476.

Seastedt, T.R., Hayes, D.C., and Petersen, N.J. 1986. Effects of vegetation, burning, and mowing

on soil macroarthropods of tallgrass prairie. In The prairie: past, present, and future : proceedings of the ninth North American prairie conference. Edited by G.K.. Clambey and R.. Pemble.

Seastedt, T.R., and Reddy, M.V. 1991. Fire, mowing and insecticide effects on soil

Sternorrhyncha (Homoptera) densities in tallgrass prairie. Journal of the Kansas Entomological Society, 64: 238–242.

Short, A.E.Z., and Fikacek, M. 2013. Molecular phylogeny, evolution and classification of the Hydrophilidae (Coleoptera). Systematic Entomology, 38: 723–752. doi:

10.1111/syen.12024.

Shorthouse, D.P. 2010. SimpleMappr, an online tool to produce publication-quality point maps.

Smart, A.J., Scott, T.K., Clay, S.A., Clay, D.E., and Ohrtman, M.K. 2013. Spring clipping, fire, and simulated increased atmospheric nitrogen deposition effects on tallgrass prairie

vegetation. Rangeland Ecology & Management, 66: 680–687.

Smith, G.W., Debinski, D.M., Scavo, N.A., Lange, C.J., John, T., Moranz, R.A., Miller, J.R., Engle, D.M., and Toth, A.L. 2016. Bee abundance and nutritional status in relation to grassland management practices in an agricultural landscape. Environmental Entomology,

45: 338–347. doi: 10.1093/ee/nvw005.

Smith, L.M., and Cherry, R. 2014. Effects of management techniques on grassland butterfly species composition and community structure. American Midland Naturalist, 172: 227–235.

Stoner, K.J.L., and Joern, A. 2004. Landscape vs. local habitat scale influences to insect

communities from tallgrass prairie remnants. Ecological Applications, 14: 1306–1320.

Sveinson, J.M.A. 2003. Restoring tallgrass prairie in southern Manitoba, Canada. [degree] thesis, University of Manitoba, .

Swengel, A.B. 1996. Effects of fire and hay management on abundance of prairie butterflies.

Biological Conservation, 76: 73–85.

Swengel, A.B. 1998. Effects of management on butterfly abundance in tallgrass prairie and pine barrens. Biological Conservation, 83: 77–89.

Swengel, A.B. 2001. A literature review of insect responses to fire, compared to other

conservation managements of open habitat. Biodiversity and Conservation, 10: 1141–1169.

Swengel, A.B., and Swengel, S.R. 2007. Benefit of permanent non-fire refugia for Lepidoptera conservation in fire-managed sites. Journal of Insect Conservation, 11: 263–279. doi: 10.1007/s10841-006-9042-9.

Swengel, S.R., Schlicht, D., Olsen, F., and Swengel, A.B. 2011. Declines of prairie butterflies in the midwestern USA. Journal of Insect Conservation, 15: 327–339. doi: 10.1007/s10841-010-9323-1.

28

Tester, J.R., and Marshall, W.H. 1961. A study of certain plant and animal interrelations on a native prairie in northwestern Minnesota. Occasional Papers of the Minnesota Museum of Natural History, 8.

Todd, T.C., James, S.W., and Seastedt, T.R. 1992. Soil invertebrate and p lant responses to mowing and carbofuran application in a North American tallgrass prairie. Plant and Soil, 144: 117–124.

Towne, E.G., and Craine, J.M. 2014. Ecological consequences of shifting the timing of burning

tallgrass prairie. PLoS ONE, 9: 1–10. doi: 10.1371/journal.pone.0103423.

Vanbergen, A.J., Woodcock, B.A., Gray, A., Grant, F., Lambdon, P., Chapman, D.S., Pywell, R.F., Heard, M.S., and Cavers, S. 2014. Grazing alters insect visitation networks and plant mating systems. Functional Ecology, 28: 178–189. doi: 10.1111/1365-2435.12191.

Vieira, L., Nascimento, P.K.S., and Leivas, F.W.T. 2018. Habitat association promotes diversity of histerid beetles (Coleoptera: Histeridae) in neotropical ecosystems. 72: 541–549.

Vogel, J.A., Debinski, D.M., Koford, R.R., and Miller, J.R. 2007. Butterfly responses to prairie restoration through fire and grazing. Biological Conservation, 140: 78–90. doi:

10.1016/j.biocon.2007.07.027.

Wagle, P., and Gowda, P.H. 2018. Tallgrass prairie responses to management practices and disturbances: a review. Agronomy, 8: 1–16. doi: 10.3390/agronomy8120300.

Wallner, A.M., Molano-Flores, B., and Dietrich, C.H. 2012. The influence of fire on Illinois h ill

prairie Auchenorrhyncha (Insecta : Hemiptera) diversity and integrity. Journal of Insect Conservation, 16: 433–445. doi: 10.1007/s10841-011-9430-7.

Warren, S.D., Scifres, C.J., and Teel, P.D. 1987. Response of Grassland arthropods to burning: a review. Agriculture, Ecosystems and Environment, 19: 105–130. doi: 10.1016/0167-

8809(87)90012-0.

Wcislo, W.T., and Cane, J.H. 1996. Floral resource utilization by solitary bees (Hymenoptera: Apoidea) and exploitation of their stored foods by natural enemies. Annual Review of Entomology, 41: 257–286.

Weaver, J.E., and Rowland, N.W. 1952. Effects of excessive natural mulch on development, yield, and structure of native grassland. Botanical Gazette, 114: 1–19.

Welti, E.A.R., Joern, A., Qiu, F., Tetreault, H.M., Ungerer, M., and Blair, J. 2019. Fire, grazing and climate shape plant–grasshopper interactions in a tallgrass prairie. Functional Ecology,

33: 735–745. doi: 10.1111/1365-2435.13272.

Wilson, J.S., and Messinger Carril, O. 2016. The bees in your backyard: a guide to North America’s bees. Princeton University Press, Princeton. p.

Wrobleski, D.W., and Kauffman, J.B. 2003. Initial effects of prescribed fire on morphology,

abundance, and phenology of forbs in big sagebrush communities in southeastern Oregon. Restoration Ecology, 11: 82–90.

Zheng, B., Cao, L., Tang, P., Achterberg, K. Van, Ho, A.A., Chen, H., Chen, X., and Wei, S.

29

2018. Molecular phylogenetics and evolution gene arrangement and sequence of mitochondrial genomes yield insights into the phylogeny and evolution of bees and sphecid wasps (Hymenoptera: Apoidea). Molecular Phylogenetics and Evolution, 124: 1–9. doi:

10.1016/j.ympev.2018.02.028.

30

Chapter 2: Spatially diverse, unmanaged sites contain greater wild bee richness and

diversity than grazed sites in the tallgrass prairie ecosystem

2.0 Abstract

Tallgrass prairie is an increasingly rare environment that houses a variety of uncommon

and vulnerable plants and animals. Prairie management includes prescribed burns and cattle

grazing, which prevent woody encroachment and support diverse plant communities. Forbs,

which drive plant diversity in the prairie, proliferate with animal pollination. Management

directed at plant communities may affect pollinators differently. A survey of wild bees was

conducted on tallgrass prairie sites that had recently experienced burns, grazing, or no

disturbance, using ground and elevated bowl traps, as well as blue vane traps. Satellite imagery

and ground surveys were used to model how landscape context affects wild bee abundance and

diversity in tallgrass prairie. Most results imply that unmanaged sites had greater abundance and

diversity of bees than grazed sites, while burned sites were often intermediate. Rarefied richness

of ground-nesting bee diversity was greatest in burned sites, but stem-nesting bee diversity had

conflicting results, with abundance and species richness significantly greater in burned sites, but

rarefied diversity negatively affected by burns. Landscape diversity, at scales around that of the

typical foraging range for solitary bees, was often part of the best models of bee abundance and

diversity for bowl captured bees. Bowl traps were better than blue vane traps at detecting effects

of disturbance type, landscape diversity, and ground cover on bee diversity. These results suggest

that grazing of tallgrass prairie should be minimized for the proper preservation of vulnerable

wild bee communities.

2.1 Introduction

Tallgrass prairie (TGP) is the most productive grassland in the Great Plains (Joyce and

Morgan 1989). Southern Manitoba represents the northernmost extent of TGP, where less than

0.05% remains due to large scale conversion to agricultural land (Joyce and Morgan 1989),

making it one of the most endangered ecosystems in North America (Samson and Knopf 1994).

TGP is home to several rare or endangered species, such as the Powesheik Skipperling, Oarisma

poweshiek (Parker), and the Western Prairie Fringed Orchid, Platanthera praeclara Sheviak &

M.L.Bowles. These rare species, and others, have helped influence conservation efforts to focus

on the possibility of long-term viability through public interest. This has largely been through

recreation and ecotourism in the TGP (Costanza et al. 1997, Veech 2006, Gerla et al. 2012,

31

Berger et al. 2020). Alarmingly, the amount of TGP in Manitoba continues to decline (Koper et

al. 2010), and decreases in precipitation due to climate change threaten to further change the

prairie (Craine et al. 2011). Studies on remnant prairie in Manitoba suggest that smaller prairie

patch sizes tend to decrease in quality over time, possibly due to increased edge effects (Koper et

al. 2010).

TGP remnants in Manitoba are managed by burning, cattle grazing, mowing (more

prevalent recently due to inability to burn because of precipitation extremes and too much soil

moisture), and, minimally, haying (Sveinson 2003, A. R. Westwood (pers. comm.) 2021). These

practices aim to simulate the natural and Indigenous set fires and ungulate grazing that created

and maintained the TGP (Pyne 1986). These practices are typically focussed on plant diversity,

and have been much better studied in more southerly American TGP remnants, but animals are

also critical for the ecological function of TGP (Knapp et al. 1999). Bees are important

pollinators in grasslands that depend directly on floral diversity (Roulston and Goodell 2011).

Suppression of naturally occurring fire and grazing can lead to dominance of C4 grasses and loss

of forb diversity. Forbs are crucial for wild bee abundance, and bumble bee diversity in the

tallgrass prairie (Howe 1994, 2011, Hines and Hendrix 2005, Kelleher and Choi 2020). Burning

the prairie reduces the litter layer, which leads to increased soil temperatures, and subsequently

earlier plant growth (Ehrenreich and Aikman 1963). However, in Kansas TGP, topography may

play a role in plant biomass, with lowland sites accumulating significantly greater live biomass

compared to upland sites in a 10-year study (Abrams et al. 1986). As well, forb and woody plants

had 200–300% greater biomass in fire free sites compared to those that were annually burned

(Abrams et al. 1986). In more easterly TGP, increasing fire frequency has been shown to

increase the species richness of summer blooming forbs (Bowles and Jones 2013). Increased soil

temperature has been shown to lead to greater nesting rates in tallgrass prairie (Buckles and

Harmon-Threatt 2019). Fire also extends the flowering season in grasslands (Mola and Williams

2018).

In contrast, grazing is known to have several detrimental effects on prairie vegetation.

Cattle grazing leads to soil compaction, and favours some exotic species over native grasses

(Ehrenreich and Aikman 1963, Walters and Martin 2003). However, both cattle and bison

grazing increase forb cover in TGP (Towne et al. 2005). Forbs are the main driver of plant

32

biodiversity in TGP (Whiles and Charlton 2006). Grazers increase the nitrogen content of the

soil and preferentially feed on grasses over forbs, which can lead to increased plant nutrient

quality, reduced grass cover, and more prominent forbs (Moran 2014). Conversely, overgrazing

has been shown to reduce soil moisture, nitrogen, and organic matter in a fertile South Dakota

grassland (Beebe and Hoffman 1968). A study in Kansas TGP found that increased native plant

diversity was best promoted using high stocking rate intensities (Hickman et al. 2004). Similarly,

in Iowa and Missouri, combination fire and cattle grazing management has been shown to restore

exotic plant laden grasslands to become similar to TGP remnants (Delaney et al. 2016).

However, changes in management practices are known to affect relative frequencies of moisture

related forb communities in TGP, with a change from haying to burning corresponding to a shift

from more dry adapted plants to increasing frequency of mesic species (Dornbush 2004). A more

recent, large scale study examining the effects of burning versus grazing on Minnesota TGP

plant and soil characteristics found that management type did not affect vegetation communities,

and grazed sites differed from burned sites primarily due to increased soil density and soil

nitrogen content (Larson et al. 2020). Overall, a combination of infrequent burning and grazing

has been shown to maximize plant diversity in TGP (Collins and Calabrese 2012), but this is not

universally implemented (Wright Morton et al. 2010). This combination does not however

appear to be beneficial to wild bees in TGP (Buckles and Harmon-Threatt 2019).

Bees forage from a centralized nest, which make them ideal bioindicators for the quality

of their local environment (Gathmann and Tscharntke 2002, Schindler et al. 2013). Furthermore,

bees have a high genetic load, making them particularly sensitive to habitat degradation (Zayed

and Packer 2005). In the short-term, fire is predicted to negatively affect stem-nesting bees and

ground-nesting bees within 10 cm of the soil surface (Cane and Neff 2011). Due to the nature of

grassland fires, however, soils as close to the surface as 2 cm can remain at moderate

temperatures during wildfire events unless conditions allow the wildfire to remain in place for

some time (Neary and Leonard 2020). Burned sites experience a prolonged flowering period and

increased amount of bare soil and dead wood for nesting (Wrobleski and Kauffman 2003, New

2014, Mola and Williams 2018). Previous management and fuel load on the prairie can however

affect the community composition of the recovering plant species following a wildfire (Ewing

and Engle 1988). Grazing could affect the ability of ground-nesting bees to successfully nest due

to increased soil compaction (Trimble and Mendel 1995), and cattle may consume some floral

33

resources used by bees (Debano et al. 2016). Several recent studies have examined the response

of bees to management practices and landscape characteristics in American TGP. Studies in

Minnesota and Iowa/Missouri found that management type did not affect bee abundance (Smith

et al. 2016, Pennarola 2019), or richness and diversity (Pennarola 2019) but demonstrated that

decreased floral abundance and increased cattle stocking rate did negatively nutritional status of

bees and had marginally significant effects on abundance (Smith et al. 2016). Buckles and

Harmon-Threatt (2019) found that grazing in TGP negatively impacted the soil and floral

resources required for bees, and bee abundance and richness trended negatively (non-

significantly) under grazing compared to burning. Most recently, Griffin et al. (2021) found that

bison grazing had a negative effect on TGP bee abundance.

Bees are diverse in semi-natural landscapes (Le Feon et al. 2010), and respond negatively

to agricultural (Connelly et al. 2015, Harrison et al. 2017) and urban intensification (Harrison et

al. 2017). Bees respond positively to increased habitat heterogeneity in TGP (Denning and Foster

2018). Wooded area in the landscape surround TGP positively affects bee communities (Griffin

et al. 2017, 2021). However, bees respond differently to management practices and landscape

fragmentation based on their life history traits such as nesting strategy (Williams et al. 2010).

There are few published studies of bees in the Canadian Prairies. Patenaude (2007) and Olynyk

(2017, Olynyk et al. 2021) examined the diversity of bees, and their response to management

respectively in mixed-grass prairie. Robson (2008) studied the structure of insect pollinator floral

visits in TGP, while Semmler (2015) studied a similar framework but with fire providing an

opportunity to test for the effect of disturbance on pollinator communities. Further afield, Evans

(2013) demonstrated that landscape context was more important than grazed status of sites on

wild bee communities in fescue prairie in Alberta (2013). Vickruck et al. (2019) examined the

benefits of wetlands to wild bees in the Alberta prairie pothole region. Similarly, further studies

in the pothole region have demonstrated that prairie reclamation from agriculture in these areas

can return bee communities back to an undisturbed state (Purvis et al. 2020). Also in Alberta,

Kohler et al. (2020) studied the effects of prairie ecoregion on bee community properties.

Sheffield et al. (2014) produced a large review of the bee species occurring in the prairie

ecoregion, which has greatly assisted in understanding known species and their ranges in the

prairie provinces. Further work on the xeric hypothesis of bee distribution has been conducted

using field studies from Manitoba, and have demonstrated that while short-tongued bees do not

34

fluctuate greatly between different prairie/moisture level types, long-tongued bees are much

more abundant/important pollinators of the forb community in mixed grass prairie than in the

relatively wet tallgrass prairie (Robson et al. 2019).

I conducted a study in the Manitoba Tall Grass Prairie Preserve (MTGPP) that had three

key objectives, 1) determine the effects of disturbance type (burning vs. grazing vs. no recent

disturbance) on abundance, richness, and diversity of bees in TGP, 2) examine the effect of

landscape and ground cover characteristics on bee abundance, richness, and diversity in TGP,

and 3) compare responses of ground and stem-nesting bees to disturbance type, landscape, and

ground cover factors. I predict that disturbed sites, burned or grazed, will increase overall wild

bee diversity relative to undisturbed controls. Sites that experienced a semi-recent fire (1–6 years

post-fire) will likely have more bare ground and dead branches for ground and stem-nesting bees

to use, as well as greater forb diversity owing to fire negating the ability of few dominant plant

species taking over and reducing overall plant diversity (Towne and Kemp 2003, Howe 2011). I

expect semi-recent cattle grazing (0–11 years post-graze) to remove dominant C4 grasses leading

to more floral diversity for bees to use to have greater plant species richness (Hickman et al.

2004, Veen et al. 2008, Wagle and Gowda 2018). Bees often have greater or similar diversity in

burned sites than grazed ones (Buckles and Harmon-Threatt 2019, Pennarola 2019), so I predict

a similar pattern in my sites. I expect landscape heterogeneity to increase bee diversity across

disturbance types (Denning and Foster 2018). Although stem-nesters are susceptible to fire, the

time since fire in my systems should not lead to differences in ground and stem-nesting bees in

burned sites (Williams et al. 2010). Griffin et al. (2017) demonstrated that bee communities on

restored prairies reach an equilibrium within 2–3 years of seeding compared to remnant prairies.

Thus, the length of time since the fire in most of my sites should have little effect. However, a

study in Minnesota found that the frequent burns that characterize early restorations of TGP have

a randomizing effect on the associated bee communities (Tonietto et al. 2017). Buckles and

Harmon-Threatt (2019) found that grazing negatively affected the soil characteristics essential

for ground nesting bees, namely moisture content, compactness, and temperature. Stem-nesters

may be positively affected by the extent of forested habitat in the landscape (Main et al. 2019).

Decreases in several pollinator populations worldwide have been well documented (Potts et al.

2010, Jacobson et al. 2018, Mathiasson and Rehan 2019). As such, I hope that this study gives

35

insight into steps that policy makers can take that safeguard bee populations, and the grasslands

that depend on their pollination services.

2.2 Methods

2.2.1 Site selection

The study was conducted at 18 sites within the Manitoba Tallgrass Prairie Preserve

(Figure 2.1), Rural Municipality of Stuartburn, in southeastern Manitoba. MTGPP is the largest

remnant TGP in Canada. The preserve is divided into a south block 8 km north o f the

U.S./Canada border, and a much wetter north block, which is bounded to the north by a large

swamp, and to the west by the Roseau River. This study took place solely in the north block. The

MTGPP exists in an agriculture/aspen forest matrix and contains over 2,000 ha of remnant TGP.

Sites were selected in consultation with Nature Conservancy Canada based on fire and

grazing history, and ease of access. The minimum distance between sites ranged from

approximately 700–1,800 m. Typical foraging ranges for solitary bees are between 150 and 600

m (Gathmann and Tscharntke 2002) with maximum foraging ranges not greatly exceeding 1 km

(Zurbuchen et al. 2010). Sites were divided into three disturbance type groups. i) Six grazed sites

were used as cattle pasture at least once since 2007 but not burned since 1993 . One site was

grazed in 2006 and 2007; one site was grazed every year between 2008–2011; two sites were

grazed in 2011 and 2012, and two sites were grazed every year 2012–2016. One site was grazed

in 2018, with an enclosure placed around my sampling area to prevent disturbance. ii) Five

burned sites that were not used for cattle grazing since at least 1993. Four sites experienced a

2012 wildfire burn and one had a controlled burn in 2016. Drought conditions during the study

period prevented the use of controlled burns, so wildfire sites were used as a proxy for fire

disturbance. iii) Seven unmanaged sites that were not grazed or burned since at least 1993. I

grouped sites of each category into six separate blocks for sampling (one block had an extra

unmanaged site and no burned site due to fewer available burned sites). Site names, disturbance

type, and coordinates can be found in the supplementary material (Table 2.9).

36

Figure 2.1 Location of the study sites showing the location of the Manitoba Tallgrass Prairie

Preserve in context to Southern Manitoba and the adjacent province and states, with an insert of the layout of the locations of the 18 study sites and their disturbance grouping, with an insert showing the design set up of each study site, with yellow circles representing bee bowl stands, and blue triangles representing blue vane traps. Maps were made using ‘SimpleMappr’

(Shorthouse 2010).

2.2.2 Sampling Procedure

In each site, a 3 x 3 array of collection stations were set up, minimally spaced 15 m apart

(Figure 2.1). Collecting stations each include a pole with a central wooden platform holding

three bee bowl traps. The platform was raised to a height of approximately 1 m. Three additional

bowls were placed on the ground at the base of each pole, for a total of 54 bowls per site (Figure

2.2). A combination of ground-level and elevated traps is effective for catching a variety of wild

bees in tallgrass prairie (Geroff et al. 2014). Pre-painted 3.5 oz bowls were purchased from New

Horizons Supportive Services Inc. (Upper Marlboro, Maryland, USA), in three colours white,

30 m N 49.18

N 49.10

W 96.74 W 96.65

Provincial Rd 201

37

fluorescent blue and fluorescent yellow. These colours are commonly used in bee surveys

(Leong and Thorp 1999).

Bee bowls were ¾ filled with a 1:1 mixture of propylene glycol and water. Traps were

deployed for 48 h. A block of three sites were sampled simultaneously. Collections were made

every two to four weeks. Three samples were taken in 2017 maintaining the order of sampling

blocks in each sample (6–29 July, 1–13 August, and 17 August-5 September). Four samples

were taken in 2018.

Figure 2.2 (Left) Example of the ground level and elevated bee bowl station. Bowls placed in the foreground for the purpose of the photograph. (Right) Photo of a blue vane trap showing the

collecting chamber (yellow) and the trapping vanes (blue) (photo courtesy of Emily Hanuschuk).

(12–22 May, 3–11 June, 3–21 July, and 15–20 August). In 2018, four blue vane traps (BVTs)

(SpringStar Inc.) were deployed spaced equidistant from neighbouring bowl stations (Figure 2.1,

2.2), BVTs were filled with a few centimetres of the propylene glycol/water mixture. BVTs have

38

been shown to be very effective at catching wild bees (Stephen and Rao 2005), and catch many

large-bodied bees in greater numbers than bee bowls (Geroff et al. 2014, Joshi et al. 2015). BVTs

were only used in the last two rounds of sampling to avoid catching bumble bee gynes. In total

there were 126 observations for bee bowl caught bees, and 36 for BVTs.

2.2.3 Processing Bees

Bees were stored in the propylene glycol/water mixture and placed in the freezer until

processing. Bees were then washed and dried following the guidelines of Droege (2008). Bees

were placed in a detergent bath and mixed using a stirring rod before being transferred to 70%

ethanol. Bees were dried using a combination of a hair dryer and paper towel. Bees were th en

pinned, labeled, and databased. All vouchers are deposited at the J.B. Wallis/R.E. Roughley

Museum of Entomology at the University of Manitoba.

Bees were identified to genus (Packer et al. 2007) and then species using published

taxonomic revisions and keys (LaBerge 1956a, 1956b, 1961, Laverty and Harder 1988, Gibbs

2011), and keys hosted on DiscoverLife.org (Andrus and Droege 2018a, 2018b, 2018c, Griswold

et al. 2018, Droege & Rehan 2019). Joel Gardner and Amber Bass (University of Manitoba)

assisted with Lasioglossum and Melissodes identifications, respectively. Bee species were

categorized based on their nesting habitat (ground vs. stem nesting) using published literature

(Sheffield et al. 2014, Gibbs et al. 2017) to test for differing guild responses to disturbance type,

landscape, and ground cover features.

2.2.4 Ground cover characteristics

To account for local effects of floral diversity on my bee captures, ground cover was

assessed twice in 2018 at each site, during the second and fourth round of bee sampling. The

average of the two assessments was used as a measure of local floral diversity in models. A 1 x 1

m quadrat was thrown haphazardly forward a few metres into the sampling area 16 times. The

proportions of non-forb cover, flowering shrub and forb cover, and bare ground were estimated

based on ground cover percentages illustrated in Folk (1951). Woody shrubs and non-woody

forbs were all included in the percent shrub/forb value. The non-forb category included grasses

and vascular plants that do not flower such as Equisetum spp. The average percent covers of the

16 replicates were taken. The number of inflorescences were recorded. Then, a principal

39

coordinate analysis was used to extract the first axis using the ‘res.pca’ function of the package

‘FactoMineR’ (Le et al. 2008) ) (Figure 2.11).

2.2.5 Landscape mapping

Landscape data for MTGPP managed sites was provided by Nature Conservancy Canada

and its partners through a licensing agreement. Landcover within a 2 km radius of the study sites

that was not included in the NCC data were hand-digitized based on satellite imagery at a

1:5,000 scale in ArcMAP v. 10.6 (ESRI 2018). Buffers with radii of 250, 500, 1000, 1500, and

2000 m were created around the centroid of the study sites. The identity of the polygons present

in the NCC data layer were simplified to correspond to readily observable features in the satellite

imagery, thus prairie subcategories were all merged as ‘prairie’, marshes, wetlands, and swamps

were all merged in to one category, and different categories of forest were merged. The

following categories were used to represent different polygons: forest, peri-urban, structure,

savannah, water, swamp, prairie, pasture, crop, river, roadside ditch (‘edge’ in Figure 2.12), road,

and other.

Each buffer for each site was intersected to extract the polygons within the radius. I used

Patch Analyst v. 5.2 (Rempel et al. 1999) to calculate landscape metrics, including the Shannon

diversity index and the mean shape index. The Shannon diversity index is a measure of patch

diversity. Shannon diversity increases as the number and proportion of different patches increase

in the buffer and is maximized when all patch types are of equal proportions (Rempel et al. 1999,

McGaril & Marks 1994). The landscape mean shape index is a measure of patch shape

complexity. It has a value of one when all patches are perfectly compact (circular in shape when

using vector data) and increases as patch shape becomes more complex. The percent of different

covers were also extracted, and then submitted to principal coordinate analysis to extract the first

axis of ‘landcover’ at different buffer ranges.

2.3 Data analyses

2.3.1 Bee diversity & community composition

All analyses were performed in R (R Core Team 2017). Spatial autocorrelation was

checked in R using the ‘DHARMa’ package (R Core Team 2017, Hartig and Lohse 2020). All

bee bowl data were combined into a site x species matrix. BVT captures were analyzed

separately. I pooled samples from a given year and included communities from both years in

40

analyses. I expected that additional species collected in spring and possibly reduced captures

from competition with blue vane traps in the summer would affect the communities recorded in

2018. Community composition was compared across sites using distance-based redundancy

analysis (dbRDA) with the Bray-Curtis index in the package vegan (Legendre and Anderson

1999, Oksanen et al. 2018). The distance-based redundancy analyses were applied with

‘Disturbance Type’ and ‘Year’ as constraints and summarized, then visualised with the ‘ordiplot’

function of the vegan package (Oksanen et al. 2018). Permutational multivariate analysis of

variance (perMANOVA) was used to test for effects of disturbance type on bee communities

using the Bray-Curtis index in the ‘adonis’ function of the vegan package (Oksanen et al. 2018).

If a significant result (p<0.05) was received for the effect of disturbance type on bee community

distances, a post-hoc test was performed using the ‘pairwise.adonis’ function of the

pairwiseAdonis package (Martinez Arbizu 2017). The betadispersion among disturbance types

was also tested using the ‘betadisper’ function of the vegan package (Oksanen et al. 2018) to

ensure that disturbance types were equally dispersed so as not to violate the assumption of equal

dispersion of perMANOVA.

Rarefied species curves were made with the iNEXT package (Hsieh et al. 2019) to

compare species richness and other diversity indices corrected to the least abundant disturbance

type. This allows a visualization of the true and extrapolated diversity at a given abundance, and

corrects for the unequal sample sizes of the disturbance types (Chao et al. 2014). I used Hill

numbers, which are measures of the effective number of species in each disturbance type (Chao

et al. 2014). These are a unified group of indices based on the Rényi entropy (Rényi 1961) and

put forth by Hill (Hill 1973). The Hill numbers include q=0, which is simply the number of

species, q=1 which is the Hill’s Shannon number of effective species, it takes abundance into

account, but still weighs rare species quite heavily compared to q=2, the inverse Simpson’s

number of effective species, which puts more weight towards more abundant species (Chao et al.

2014). To understand how Hill numbers correspond to rarefied species curves, it is useful to note

that the Shannon and Simpson’s curves produce what is essentially an “effective species

number”, which is based on comparing the number of species recovered given the abundance of

the sample and translating that to give the number of species required to produce that same

number if all species were equally abundant (Chao et al. 2014). As noted by Chao et al. (2014),

the use of these Hill numbers in biodiversity studies solves many problems that have been raised

41

in the literature. A notable improvement is that these Hill values behave more intuitively. For

example, they double when two communities are combined that have no species in common but

the same abundance distribution.

2.3.2 Modelling

Generalized linear mixed-effects models (package glmmADMB (Skaug et al. 2012))

were used to test for differences between abundance and diversity (species richness and the

Hill’s Shannon number of effective species) among disturbance types (Jost 2007, Fournier et al.

2012, Skaug et al. 2012). I examined several explanatory variables including, year and day-of-

year. Meteorological data from a nearby weather station (Emerson MB, Environment and

Climate Change Canada) were included (average temperature and average precipitation of the

days the traps were out). Landcover data were included in a PCA and the first axis used in

models (Figure 2.12). The first axis of a ground cover PCA was also used in the models. Site was

used as a random effect in all models.

We first ran five global models that differed only by buffer size of the landscape data and

decided to analyse all models using the model with the lowest Akaike Information Criterion with

small sample size correction (AICc) score for bee abundance, which was the 1 km buffer. I then

created a model that included only year, BVT presence, average temperature, average

precipitation, and day of year. I then used the ‘dredge’ function of the MuMin package (Barton

2018) to see which of these variables appeared important based on their presence or absence in

the model with the lowest AICc. The variables that were part of the best model (based on lowest

AICc) using only those factors were kept in my base model for additional analyses. The

landscape variables, landcover PCA 1st axis, landscape mean shape index, landscape Shannon

diversity, groundcover PCA 1st axis, and disturbance type were added to the base model. The

model was then dredged again, and the model with the lowest AICc was chosen as the best

model.

Because all response variables were positive numbers, for discrete models, the negative

binomial, quasi-Poisson, or Poisson distributions were used based on which had the lower AICc

score and lower dispersion parameter (for negative binomial vs. quasi-poisson), or based on

whether or not they were overdispersed (Poisson) using the ‘check_overdispersion’ function of

the ‘performance’ package (Ludecke et al. 2020) (using the equivalent glmmTMB model). For

42

continuous models, the gamma distribution was used, except for the blue vane species richness

model, which fit a gaussian distribution. Models were checked for collinearity in their

explanatory variables, and their residual plots examined by using the ‘check_model’ function of

the ‘performance’ package (Ludecke et al. 2020), with correlations greater than (-) 0.75

considered highly correlated. The models and their alternate distributions were also ranked using

the ‘compare_performance’ function, but their dispersion parameters were the deciding factor if

negative binomial and quasi-Poisson distributions were approximately equivalent.

2.4 Results

2.4.1. Bee captures

I collected 1,562 bees in bee bowls over two years, representing 76 species in 20 genera

(Table 2.6, 2.7). The most efficient bee bowls, based on number of captures, were blue ground

bowls, and the least efficient were yellow elevated bowls. The majority of bee captures belonged

to the genus Lasioglossum (955 specimens), mostly in the subgenus Dialictus. I found one

provincial record, Lasioglossum (Dialictus) michiganense (Mitchell), a rare social parasite on

other Dialictus. The earlier start in the second field season allowed for more representation from

typically spring flying genera such as Osmia and Andrena. Blue vane traps caught 2,040 bees

representing 45 species in 17 genera (Table 2.8). Blue vane captures were dominated by

Melissodes spp. (1,238 specimens) and Bombus spp. (321 specimens). One bee species in the

blue vane captures, Triepeolus helianthi (Robertson), is previously unrecorded from the

province. One specimen of the oil collecting bee, Macropis nuda (Provancher), was found in the

blue vane trap captures.

2.4.2 Community Composition - Distance-based Redundancy Analysis & perMANOVA

Bee bowl community composition was significantly affected by disturbance history

(Figure 2.3, Table 2.1). Burned sites differed from grazed ones, but neither differed from

undisturbed sites. No difference in community composition was found when blue vane trap data

were analyzed. Disturbance and year explained approximately 19% of the variation in the total

bee communities, and approximately 16% of the variation in the ground nesting bee

communities. Burned and grazed bee communities may be marginally different when all bee

bowl captures are considered together.

43

Figure 2.3 Ordination plots of the Bray-Curtis distance-based redundancy analysis of bee bowl community data (all bees, left; ground-

nesting bees, centre; stem nesting bee communities, right) constrained by disturbance type and year. Each point represents a site in a

given year (2017 or 2018). Disturbance type had a significant effect on the combined community composition in both the

perMANOVA (Table 2.2, F = 1.75, R2 = 0.09, p = 0.017) and dbRDA (F = 1.6548, p = 0.034), resulting from a difference between

grazed and burned sites according to the perMANOVA (Tukey’s post-hoc test with a Bonferroni correction p = 0.078). In the ground

nesting bee communities, disturbance type had a significant effect according to perMANOVA (Table 2.2, , F = 1.76, R 2 = 0.09, p =

0.032) and dbRDA (F = 1.6280, p = 0.049), however a post-hoc test on the perMANOVA results did not uncover a significant

difference between any two disturbance types. Stem-nesting bee communities were not found to be significantly different by

disturbance type in either perMANOVA or dbRDA.

44

Figure 2.4 Ordination plot of the Bray-Curtis distance-based redundancy analysis of blue vane

trap captured bee communities constrained by disturbance type. Samples were combined by site

to avoid pseudo-replication. perMANOVA on the Bray-Curtis distances between disturbance

types revealed that they were not significantly distant from each other (p = 0.962) (Table 2.2).

Similarly, dbRDA did not uncover a significant effect of disturbance type.

Table 2.1 Summary of perMANOVAs performed on the Bray-Curtis distance of different bee communities. The Post-hoc column represents the results of a Tukey’s post-hoc test with a Bonferroni correction. B = ‘Burned’, G = ‘Grazed’, U = ‘Undisturbed’.

Community F model R2 p value Post-hoc

All bee bowl captures

1.7489 0.08763 0.017 B-G: 0.078 B-U: 0.291 G-U: 0.546

Blue vane trap captures

0.429 0.05411 0.962

Ground-nesting bee bowl captures

1.7565 0.09052 0.032 B-G: 0.141 B-U: 0.198

G-U: 0.720 Stem-nesting bee

bowl captures

1.3924 0.06548 0.126

2.4.3. Abundance, Species Richness, and Hill’s Shannon Diversity Modeling

45

2.4.3.1. Effect of Disturbance Type

All models used land cover variables at the 1 km radius based on that distance producing

the lowest AICc score in my global abundance model. Disturbance type was never among the

best models for blue vane trap captures. However, my best-fit model for bee bowl abundance

found a significant effect of disturbance type. Data were over-dispersed, so a negative binomial

distribution was used. Grazed sites supported significantly fewer bees than undisturbed sites

(Figure 2.5). Similarly, disturbance type was a significant variable in my best fitting model for

species richness (Figure 2.5). The data were over-dispersed, and a quasi-Poisson distribution

resulted in a lower dispersion parameter than a negative binomial distribution. In the case of

species richness, grazed sites were less species rich than both burned and undisturbed sites.

Disturbance type was also recovered in the best model to explain Hill’s Shannon effective

species number (Figure 2.5). The effective species number model was fit with a gamma

(link=log) distribution. Once again, grazed sites supported less diversity than undisturbed sites.

For ground nesting bees caught in bee bowls, disturbance type was not part of the best

model for bee abundance (Table 2.2). However, disturbance type was a signif icant variable in the

model that best explained ground nesting bee species richness (Figure 2.6). The model was

overdispersed and conformed approximately to a quasi-Poisson distribution (Table 2.3). Grazed

sites had significantly less species richness than undisturbed sites, however ground nesting bee

Hill’s Shannon (q=1) diversity was not significantly affected by disturbance type (Table 2.3).

46

Figure 2.5 Effect plot showing the predicted effect of disturbance type on bee bowl trap abundance (left), species richness (centre),

and Hill’s Shannon effective species number (right) when all other variables in the GLMM are held at their average or default state.

Different letters above the disturbance types represent significantly different results (p = less than 0.05) based on a GLMM where site

was used as a random effect. Points represent the median value of the predictor variable of the disturbance type subset , and whiskers

represent 95% confidence intervals.

47

Stem-nesting bee abundance was significantly affected by disturbance type, with burned

sites supporting significantly more stem nesting bees than grazed sites (Figure 2.7). The model

was over-dispersed and fit with a negative binomial distribution. Stem-nesting bee species

richness was also affected by disturbance type, with burned sites having greater species richness

than grazed sites (Figure 2.8). Stem-nesting bee effective species numbers were not significantly

different between disturbance types (Table 2.4).

Figure 2.6 Effect plot showing the predicted effect of disturbance type on ground nesting bee

species richness when all other variables in the GLMM are held at their average or default state.

Different letters above the disturbance types represent significantly different results (p = less

than 0.05) based on a GLMM where site was used as a random effect. Points represent the

median value of the predictor variable of the disturbance type subset, and whiskers represent

95% confidence intervals.

48

49

Figure 2.7 Effect plot showing the predicted

effect of disturbance type on stem nesting bee

abundance when all other variables in the GLMM

are held at their average or default state. Different

letters above the disturbance types represent

significantly different results (p = less than 0.05)

based on a GLMM where site was used as a

random effect. Points represent the median value

of the predictor variable of the disturbance type

subset, and whiskers represent 95% confidence

intervals.

Figure 2.8 Effect plot showing the predicted

effect of disturbance type on stem nesting bee

species richness when all other variables in the

GLMM are held at their average or default state.

Different letters above the disturbance types

represent significantly different results (p = less

than 0.05) based on a GLMM where site was used

as a random effect. Points represent the median

value of the predictor variable of the disturbance

type subset, and whiskers represent 95%

confidence intervals.

50

2.4.3.2. Effect of landscape characteristics

Of the three landscape characteristics I analysed, the landscape Shannon diversity and

the landcover PCA appeared most consistently in the best models for bee bowl trap captures . The

landcover PCA had several landcover types along the primary axis, including prairie, water,

savannah, and swamp grouping along the negative side of the first axis, and forest, pasture, ditch,

and road on the positive side of the first axis. The first axis of the landcover PCA accounted for

over 40% of the variation in the measured landcover types. Both Shannon diversity and the

landcover PCA had a positive effect on abundance, richness, and Hill’s Shannon effective

species number, implying that more diverse landscapes support greater wild bee diversity in the

prairie, and that increased cover of variables like forest and ditch (‘edge’ in Figure 2.12) had a

positive effect, while increased prairie, savannah, and swamp cover had a negative effect on bee

diversity. The mean shape index, which approximates the amount of edge in the environment,

also had a positive effect on bee effective species number for bee bowl captured bees. The trend

was the same when looking at only the ground nesting captures, however only the stem nesting

bee abundance and richness responded significantly to increasing landscape Shannon diversity.

The blue vane trap captures did not pick up on any landscape effects, except for a

positive effect of the landcover PCA on the number of effective species.

2.4.3.3. Effect of groundcover characteristics

The combined bee bowl captures, as well as the ground nesting captures responded to the

groundcover PCA in several models. The groundcover PCA had percent grass cover positioned

on the negative aspect of the first axis, while percent forbs, number of flowers, and to a lesser

extent percent bare ground grouped on the positive side of the first axis. The first axis of the

groundcover PCA accounted for over 64% of the variation in the measured ground

characteristics. In all cases, the groundcover PCA had a positive effect, which goes in the

direction of increased forb cover, number of flowers, and slightly towards more increased bare

ground in the environment, and away from increasing grass cover (Figure 2.11). Neither stem

nesting bees, nor blue vane captures responded to the groundcover PCA.

2.4.4 Rarefaction

Grazed sites had significantly less rarefied diversity for each of the first three Hill

numbers. This pattern was less evident in blue vane captures, which only showed an effect

51

between grazed (less diverse) and undisturbed (more diverse) sites when q=2 and dominant

species are given greater weight in the analysis. Ground-nesters and stem-nesters differ in their

rarefied diversity when q=1 and 2. Ground-nesters are like the total community and accumulate

diversity more rapidly in burned sites. But stem-nesters show the greatest rarefied diversity in

undisturbed sites and burned sites are lowest, though this may be driven by high abundance of

common species in burned sites lowering the diversity (Figure 2.9, Figure 2.10).

52

Figure 2.9 Rarefied curves of the entire bee bowl captured community (top), and the blue vane captured community (bottom). The Q

numbers represent the different Hill numbers, with q=0 representing rarefied species richness, q=1 representing rarefied Hill’s

Shannon effective species number, and q=2 representing rarefied Inverse Simpson’s effective species number.

S

pec

ies

Div

ersi

ty

Q=0 Q=1

Q=2

S

pec

ies

Div

ersi

ty

Q=0 Q=1

Q=2

53

Figure 2.10 Rarefied curves of the ground nesting bee bowl captured community (top), and the stem nesting bee bowl community

(bottom). The Q numbers represent the different Hill numbers, with q=0 representing rarefied species richness, q=1 representing

rarefied Hill’s Shannon effective species number, and q=2 representing rarefied Inverse Simpson’s effective species number.

S

pec

ies

Div

ersi

ty

S

pec

ies

Div

ersi

ty

Q=0

Q=0

Q=1

Q=1

Q=2

Q=2

54

2.5 Discussion

I found significantly greater abundance and diversity in undisturbed sites of the TGP than

in grazed sites. Burned sites were intermediate. Pollinator abundance is positively correlated with

TGP plant diversity (Kelleher and Choi 2020). Floral richness in TGP is maximized under

disturbance at intermediate timescales (burning only once every four years) in Kansas TGP, but

too frequent disturbances act as extinction causing events (Collins et al. 1995). Alternatively, the

litter accumulation in undisturbed sites itself acts as a form of disturbance, leading to increased

plant diversity (Bascompte and Rodríguez 2000). The latter theory could explain the increased

abundance and diversity of bees on my undisturbed sites compared to my grazed sites if they are

directly responding to greater floral richness, although it is not entirely clear if greater floral

diversity predicts wild bee status in TGP (Griffin et al. 2021). Conversely, bumble bees are

known to be more diverse in TGP sites with greater floral diversity (Hines and Hendrix 2005).

These conflicting results could be because increased floral diversity does not necessarily equate

to benefits for bees if the increased richness comes at the cost of less coverage of a more

rewarding abundant floral source. As well, not all floral resources can be used by bees in TGP

due to floral morphology, for example the Western Prairie Fringed Orchid, which is adapted

solely for moth pollinators. The greater rarefied richness of bees in my burned, or even the

greater abundance of bees in my undisturbed sites, could be in response to these mechanisms.

Hymenopteran pollinators have been shown to respond positively, via increased abundance and

species richness, to wildfire events (Carbone et al. 2019), which characterized the majority of my

burned sites. Pollinator communities are known to show greatest abundance and diversity under

wild, but infrequent (10-12 year gaps) (Carbone et al. 2019), high intensity fires (Galbraith et al.

2019). Prescribed fires do not reflect the real intensity of once more common tallgrass prairie

fires, which may be artificially affecting studies on prairie ecosystem functioning (Twidwell et

al. 2016).

Though undisturbed sites may have overall higher plant diversity, woody encroachment,

a potential consequence of fire suppression, can reduce abundance and diversity of perennial

dicot forbs and legumes, many of which are likely bee pollinated (Taft and Kron 2014).

Increasing shade brought on by woody plant encroachment decreased plant-pollinator

interactions in Illinois prairie, with more shaded flowers receiving less pollen input from

pollinators, ultimately affecting their ability to propagate (Chi and Molano-Flores 2015).

55

Conversely, greater woody encroachment increased the amount of bare ground in the

environment (Taft and Kron 2014), which could explain my positive effect of forest in the

landscape on ground nesting bees.

A review on the effect of grazing on grassland plants and arthropods found that plants

were not affected by increasing grazing pressure, however arthropods were often negatively

affected by increased grazing intensity (van Klink et al. 2015). Even relatively infrequent

trampling events on soil can negatively affect grassland arthropods, much less than what is

required to affect grassland plant communities (Duffey 1975, van Klink et al. 2015). Grazing is

known to compact the soil in grassland environments (Trimble and Mendel 1995). This could

make it more difficult for ground nesting bees to dig nests, though as Sardiñas and Kremen

(2014) point out, compaction preference is likely context dependent. It also reduces water

infiltration (Trimble and Mendel 1995), which could presumably have a positive effect if there is

no standing water, or a very detrimental effect on ground nesting bees if water is allowed to

collect above nest entrances, which was very likely the case in all of my sites, where the water

table was at the surface of the soil in the North block of the MTGPP, where I was sampling, until

early July (personal observation). This despite the fact that early season conditions in both 2017

and 2018 were drier than usual in Southern Manitoba (Manitoba Conservation and Water

Stewardship 2017, 2018). This could also explain why increasing road and ditch cover in the

environment, which are well drained, increased bee abundance, richness, and diversity in my

combined and ground nesting bee models.

This study supports a few recent studies that point to the negative effects that grazing has

on wild bee populations in tallgrass prairie (Buckles and Harmon-Threatt 2019, Griffin et al.

2021). Another study compared TGP specifically with grazed bison and cattle pastures and found

that restored TGP housed a greater abundance and richness of wild bumble bees, however they

unfortunately did not disclose how or if the TGP in their study was actively managed

(Rosenberger and Conforti 2020). Some grassland studies have found positive effects of grazing

on bee abundance and species richness compared to abandoned and undisturbed sites (Kormann

et al. 2015), however they did find a negative effect of grazing on red-listed species richness

across multiple taxa. In grassland sites exposed to differing levels of number of mowing regimes,

wild bee richness, the number of insect-flower interactions, and plant species richness were all

56

greater in sites with a history of lower intensity mowing (Weiner et al. 2011), which could

explain my grazing effects if they are acting equivalently and occurring too frequently or under

too much intensity. Unfortunately, grazing intensity records at my sites were not available.

Stem-nesting bees had lower rarefied diversity in my burned sites, though greater Hill’s

Shannon diversity in my models. This seemingly conflicting result is likely the result of a high

abundance of Hoplitis pilosifrons (Cresson 1864) in burned sites lowering the rarefied diversity.

Above-ground nesting bees are more susceptible to nest destruction from disturbances than

below-ground nesting species (Spiesman et al. 2019). Insects tracked in a burned TGP recovered

within two years, however inability to move away from the fire, such as the case of larval and

pupal stem nesting bees, was a significant determinant of initial negative effects (Panzer 2002).

Most of my burned sites were subjected to a high intensity fire in 2012, which could have

contributed to the high proportions of ground nesting to stem nesting bees found in this study, as

well as a previous study on the effects of wildfire in MTGPP (Semmler 2015). However, a global

synthesis failed to find a difference in response to fire between abundance and richness of

ground nesting and stem nesting hymenopterans, though all hymenopterans had a more positive

response to wildfires than to prescribed fires (Carbone et al. 2019).

Decreased landscape diversity reduces wild bee species richness in grassland ecosystems

(Jauker et al. 2013, Hopfenmuller et al. 2014). Greater landscape heterogeneity in Kansas TGP

increased wild bee diversity (Denning & Forster 2018). My results suggest that bees respond

negatively to increasing prairie and savannah landcover, and benefit from greater crop and

pasture lands. This contrasts with other studies that have found that bees respond positively to

increasing grassland cover in the environment, both in the tallgrass prairie ecoregion (Denning

and Foster 2018, Evans et al. 2018, Griffin et al. 2021) and in other grassland systems (Sjödin et

al. 2008). Bees are negatively impacted by increased crop cover in the environment (Holzschuh

et al. 2010, Kohler et al. 2020, Shaw et al. 2020), but my results suggest that pasture and crop

areas in the landscape benefitted bees, but this could be due to bees being more concentrated in

my prairie patches when crop and pasture were present in the surroundings, or pasture and crop

areas, which together accounted for an average of 21.73 +/- 15.01% of cover type, may be more

well drained than the relatively wet prairie. Open water and swamp are correlated with prairie

and savannah in the landscape surrounding my study sites. Alternatively, pasture and crop area

57

may be correlated with a cover type such as ditch that is the actual source of beneficial area to

bees. However, pastures can benefit wild bees in agriculturally intensive areas (Morandin et al.

2007), and should perhaps be included in future studies in TGP given their ubiquity, at least for

the purposes of comparison to remnant sites.

The areas adjacent to prairie fields are known to influence prairie bee communities.

Positive effects of forest cover (Evans et al. 2018, Griffin et al. 2021) or forest edge (Neumüller

et al. 2020, Rivers-Moore et al. 2020) on wild bee communities in grasslands, have been found

previously, although successional stage of the forest may be an important, overlooked factor

(Odanaka and Rehan 2020). Neumüller et al. (2020) found a positive effect of ruderal areas on

wild bees, which could be equivalent to my positive effect of ditches/road verges, as both are

characterised by high numbers of dicot forbs, although I did not quantify this in my study. Prairie

road verges seeded with prairie plants are known beneficial resources for bees owing to their

floral and ground nesting opportunity resources (Hopwood 2008). Another study in Manitoba

prairies found a negative effect of proximity to road edges on wild bees, though specifically only

on social bees (Olynyk et al. 2021). One large, multi-national study confirmed that diverse

landscapes had positive effects on wild bees, but did not uncover significant effects of a

landscape variable that included a patch shape index (Kennedy et al. 2013). In contrast, one

study in TGP did find a positive effect of edge density on wild bees (Denning and Foster 2018),

consistent with my findings.

Several studies have successfully demonstrated that bees are positively affected by

increased flower or forb abundance (Sjödin et al. 2008, Crist and Peters 2014), and specifically

by native flowering plant abundance in TGP (Stein et al. 2020). It has also been shown that bee

richness is positively affected by increased flowering plant richness (Batáry et al. 2010, Lane et

al. 2020). On the other hand, other studies demonstrate a clear effect of disturbance type with no

effects of flower abundance or richness on wild bees in TGP (Davis et al. 2008, Griffin et al.

2021). My results, along with another recent grassland study (Rotchés-Ribalta et al. 2018),

suggest that both disturbance type and, though my groundcover measurements were very

preliminary, local landscape factors are at play in shaping wild bee communities. Future studies

should analyse a much greater suite of soil and grass/forb vegetation factors to determine how

these factors may be behind the conflicting results found in studies of how disturbance drives

58

prairie insect composition (A. R. Westwood (pers. comm.) 2021). Reduction in floral availability

is proposed to be one of the most important contributors to the worldwide decline of wild bees

(Goulson et al. 2015). Pollinator visitation rates are known to be maximized under low intensity

management, and pollinator species richness is likewise maximized with high flowering plant

richness and cover in grassland ecosystems (Hudewenz et al. 2012). A study on bumblebee

abundance and diversity found that both measures were influenced by local and landscape scale

features in small TGP remnants (Hines and Hendrix 2005).

Functional traits may mediate pollinator responses to management (Rotchés-Ribalta et al.

2018). I found that both ground nesting bees and stem nesting bees appear to generally respond

positively to burning, at least relative to grazing. However, rarefaction analysis suggest that

stem-nesting bees accumulate diversity more slowly in burned sites. A study on uncultivated

strips in European agricultural grasslands found that only above ground nesting bumblebees

responded positively to increasing landscape heterogeneity (Ekroos et al. 2013), which agrees

with my models. Studies in European grasslands have reported significant effects of landscape

composition and diversity on wild stem-nesting bees (Steckel et al. 2014), but I failed to see this

pattern in my data. It is possible that I simply did not capture enough stem nesting bees in my

study to generate conclusive results on the effects of landscape and groundcover in my stem

nesting bee GLMMs. Alternatively, the randomizing effect of disturbance of TGP management

on bee communities, as demonstrated by high beta diversity of the bee communities between

restored sites in Tonietto et al. (2017), could be a significant factor in shaping the bee

communities that I caught. This is explained by disturbance leading to a somewhat random

founding population occurring in sites after disturbance, with sites becoming more similar with

increasing time since disturbance (Tonietto et al. 2017).

More general pollinator assessments should include syrphid flies, as flies are more

frequent flower visitors in northern TGP (Robson 2008). My sites were in wet prairies that may

have limited nesting for bees but would benefit moisture-dependent syrphids. Indeed, a previous

study in the MTGPP found that there was increased syrphid abundance at wetter sites (Semmler

2015). Furthermore, some large syrphids carry as much pollen as honey bees, but they are too

often ignored in pollination studies (Chisausky et al. 2020). Other flies are even more neglected

59

as pollinators although collectively they may transport five times more pollen than syrphid flies

in agricultural settings (Orford et al. 2015).

Trap type can influence bee samples in TGP, but context matters (McCravy et al. 2019).

In my study, blue bowl traps were the most successful, and yellow bowl traps the least

successful, at catching bees. In contrast, yellow traps were most effective in an Iowa TGP study

(Kwaiser and Hendrix 2008). This variability is an important justification for using traps of

multiple colours in surveys. However, combining trap types may not always improve clarity.

Bowl traps showed significantly lower species richness when BVT were present in my study.

Furthermore, BVTs failed to detect effects of disturbance type, local groundcover, and landscape

level composition and configuration, which were evident in the bowl data. One study found that

BVTs are more accurate at representing captures that are driven by usable habitat in the area, but

worse at reflecting the importance of local scale factors (Rhoades et al. 2017). This somewhat

agrees with my finding that landcover mattered in the case of bee Hill’s Shannon diversity in my

BVT captures and that groundcover effects were not present in any of my BVT models. BVTs

have also been shown to bee poor proxies for net collections from targeted plant species (Gibbs

et al. 2017). This suggests that BVTs should be used with caution when assessing endangered

area biodiversity, as they may fail to pick up on significant threats to wild pollinators and may

interfere with other trapping methods.

Future studies should measure soil properties, including litter depth, as increased litter

depth is known to have negative effects on wild bees in Manitoba prairies (Olynyk et al. 2021).

This could be an important factor explaining why grazed sites were overall worse for bees, and

potentially why grazed sites showed significantly less rarefied richness than burned sites in both

the total bee and ground nesting bee communities.

I caught a single Macropis nuda female in one of my blue vane traps, and I observed their

host plant, Lysimachia at some of my sites. Two other M. nuda specimens are known from the

preserve and its immediate vicinity (Gibbs et al. 2021). Macropis nuda is a host of the federally

endangered cuckoo of this bee, Epeoloides pilosulus (Cresson), which has been recently

collected near the preserve (Gibbs et al. 2021). Additional efforts to locate this rare species in

MGTPP and determine effects of disturbance type on its survival would be warranted.

60

My study sites were relatively close together due to constraints of the MGTPP and areas

within it that had been with burns and grazing allow sufficient time to conduct this study along

with a parallel study using pitfall traps. I found no evidence of spatial autocorrelation, but it

would be beneficial to sample a wider breadth of TGP remnants in the province to see if my

results are consistent across different regions. Comparison to sites with prescribed fires would be

important to determine if my wildfire sites are useful proxies for this disturbance type.

My sites were quite depauperate and variable of flowers, as demonstrated by the average

number and standard deviation in my quadrat samples, 23.75 (±20.71), but other areas of the

MGTPP can have locally dense flower patches (pers. obs.). Bowl trapping alone may be

sufficient in sampling bee communities when floral abundance is low (Roulston et al. 2007).

However, passive sampling is worse at capturing abundance and richness of wild bees compared

to net sampling (Prendergast et al. 2020), suggesting that a follow up study should incorporate

net sampling in the TGP, at least at the relatively flower rich roadsides.

Given the conflicting results obtained among several TGP bee studies, it appears likely

that differences in burn season/intensity and grazing intensity can create quite dissimilar

conditions between studies. Therefore, as many land use history, vegetation, and soil variables as

possible should be collected during future research to try and unravel the seeming discrepancies

between studies.

2.6 Conclusion

My study shows that different disturbance practices can lead to significant changes in

TGP bee communities, and that cattle grazing should be minimally implemented to best preserve

wild bee abundance and diversity. My results also demonstrate that landscape diversity in TGP

benefits bees, as well as suggesting that broader landcover, and local ground cover influence bee

bowl captures in TGP. As well, this study highlights the potential hazards of sampling solely

with blue vane traps, as they failed to pick up on the patterns observed in the bee bowl captures.

My rarefied curves also highlighted the differential response to disturbance type between ground

nesting and stem nesting bees, implying that refugia should be maintained, when possible, to

maximize TGP bee diversity.

2.7 References

61

A. R. Westwood (pers. comm.). 2021. Personal Communication.

Abrams, M.D., Knapp, A.K., and Hulbert, L.C. 1986. A ten-year record of aboveground biomass in a Kansas tallgrass prairie: effects of fire and topographic position. American Journal of

Botany, 73: 1509–1515.

Andrus, R., and Droege, S. 2018a. Draft guide to the male Coelioxys of eastern North America. Available from https://www.discoverlife.org/mp/20q?guide=Coelioxys_male.

Andrus, R., and Droege, S. 2018b. Draft guide to the Stelis of eastern North America. Available

from https://www.discoverlife.org/mp/20q?guide=Stelis.

Andrus, R., and Droege, S. 2018c. Draft guide to the female Megachile of eastern North America. Available from https://www.discoverlife.org/mp/20q?guide=Megachile_female.

Barton, K. 2018. MuMIn: Multi-model inference.

Bascompte, J., and Rodríguez, M.A. 2000. Self-disturbance as a source of spatiotemporal heterogeneity: the case of the tallgrass prairie. Journal of Theoretical Biology, 204: 153–164. doi: 10.1006/jtbi.2000.2002.

Batáry, P., Báldi, A., Sárospataki, M., Kohler, F., Verhulst, J., Knop, E., Herzog, F. , and Kleijn,

D. 2010. Effect of conservation management on bees and insect-pollinated grassland plant communities in three European countries. Agriculture, Ecosystems and Environment, 136: 35–39. doi: 10.1016/j.agee.2009.11.004.

Beebe, J.D., and Hoffman, G.R. 1968. Effects of grazing on vegetation and soils in Southeastern

South Dakota. The American Midland Naturalist, 80: 96–110.

Berger, H., Meyer, C.K., Mummert, A., Tirado, L., Saucedo, L., Bonello, H., Arsdale, D. Van, and Williams, G. 2020. Land use dynamics within the tallgrass prairie ecosystem: the case for the Conservation Reserve Program ( CRP ). Theoretical Ecology, 13: 289–300.

Available from https://doi.org/10.1007/s12080-020-00452-z.

Bowles, M.L., and Jones, M.D. 2013. Repeated burning of eastern tallgrass prairie increases richness and diversity, stabilizing late successional vegetation. Ecological Applications, 23: 464–478. doi: 10.1890/12-0808.1.

Buckles, B.J., and Harmon-Threatt, A.N. 2019. Bee diversity in tallgrass prairies affected by management and its effects on above‐ and below‐ground resources. Journal of Applied Ecology, 56: 2443–2453. doi: 10.1111/1365-2664.13479.

Cane, J.H., and Neff, J.L. 2011. Predicted fates of ground-nesting bees in soil heated by wildfire:

thermal tolerances of life stages and a survey of nesting depths. Biological Conservation, 144: 2631–2636. doi: 10.1016/j.biocon.2011.07.019.

Carbone, L.M., Tavella, J., Pausus, J.G., and Aguilar, R. 2019. A global synthesis of fire effects on pollinators. Global Ecology and Biogeography, 28: 1487–1498. doi: 10.1111/geb.12939.

Chao, A., Gotelli, N.J., Hsieh, T.C., Sander, E.L., and Colwell, R.K. 2014. Rarefaction and extrapolation with Hill numbers : a framework for sampling and estimation in species diversity studies. Ecological Monographs, 84: 45–67.

62

Chi, K., and Molano-Flores, B. 2015. Degradation of habitat disrupts plant – pollinator interactions for a rare self-compatible plant. Plant Ecology, 216: 1275–1283. doi: 10.1007/s11258-015-0507-3.

Chisausky, J.L., Soley, N.M., Kassim, L., Bryan, C.J., Gonçalves Miranda, G.F., Gage, K.L., and Sipes, S.D. 2020. Syrphidae of Southern Illinois: diversity, floral associations, and preliminary assessment of their efficacy as pollinators. Biodiversity Data Journal, 8: e57331. doi: 10.3897/BDJ.8.e57331.

Collins, S.L., and Calabrese, L.B. 2012. Effects of fire, grazing and topographic variation on vegetation structure in tallgrass prairie. Journal of Vegetation Science, 23: 563–575. doi: 10.1111/j.1654-1103.2011.01369.x.

Collins, S.L., Glenn, S.M., and Gibson, D.J. 1995. Experimental analysis of intermediate

disturbance and initial floristic composition: decoupling cause and effect. Ecology, 76: 486–492.

Connelly, H., Poveda, K., and Loeb, G. 2015. Landscape simplification decreases wild bee pollination services to strawberry. Agriculture, Ecosystems and Environment, 211: 51–56.

doi: 10.1016/j.agee.2015.05.004.

Costanza, R., D’Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O’Neill, R. V, Paruelo, J., Raskin, R.G., Sutton, P., and van den Belt, M. 1997. The value of the world’s ecosystem services and natural capital. Nature, 387: 253–260.

Crist, T.O., and Peters, V.E. 2014. Landscape and local controls of insect biodiversity in conservation grasslands: implications for the conservation of ecosystem service providers in agricultural environments. Land, 3: 693–718. doi: 10.3390/land3030693.

Davis, J.D., Hendrix, S.D., Debinski, D.M., and Hemsley, C.J. 2008. Butterfly, bee and forb

community composition and cross-taxon incongruence in tallgrass prairie fragments. Journal of Insect Conservation, 12: 69–79. doi: 10.1007/s10841-006-9063-4.

Debano, S.J., Roof, S.M., Rowland, M.M., Smith, L.A., and Debano, S.J. 2016. Diet overlap of mammalian herbivores and native bees: implications for managing co- occurring grazers

and pollinators. Natural Areas Journal, 36: 458–477.

Delaney, J.T., Moranz, R.A., Debinski, D.M., Engle, D.M., and Miller, J.R. 2016. Exotic-dominated grasslands show signs of recovery with cattle grazing and fire. PLoS ONE, 11: e0165758. doi: 10.1371/journal.pone.0165758.

Denning, K.R., and Foster, B.L. 2018. Taxon-specific associations of tallgrass prairie flower visitors with site-scale forb communities and landscape composition and configuration. Biological Conservation, 227: 74–81. doi: 10.1016/j.biocon.2018.08.023.

Dornbush, M.E. 2004. Plant community change following fifty-years of management at Kalsow

Prairie Preserve, Iowa, U.S.A. American Midland Naturalist, 151: 241–250.

Droege, S. 2008. The very handy manual : how to catch and identify bees and manage a collection.

Duffey, E. 1975. The effects of human trampling on the fauna of grassland litter. Biological

63

Conservation, 7: 255–274.

Ehrenreich, J.H., and Aikman, J.M. 1963. An ecological study of the effect of certain management practices on native prairie in Iowa. Ecological Monographs, 33: 113–130.

Ekroos, J., Rundlöf, M., and Smith, H.G. 2013. Trait-dependent responses of flower-visiting insects to distance to semi-natural grasslands and landscape heterogeneity. Landscape Ecology, 28: 1283–1292. doi: 10.1007/s10980-013-9864-2.

Evans, E., Smart, M., Cariveau, D., and Spivak, M. 2018. Wild, native bees and managed honey

bees benefit from similar agricultural land uses. Agriculture, Ecosystems and Environment, 268: 162–170. doi: 10.1016/j.agee.2018.09.014.

Evans, M. 2013. Influences of grazing and landscape on bee pollinators and their floral resources in rough fescue prairie. [degree] thesis, University of Calgary, . doi:

10.11575/PRISM/26522.

Ewing, A.L., and Engle, D.M. 1988. Effects of late Summer fire on tallgrass prairie microclimate and community composition. The American Midland Naturalist, 120: 212–223.

Le Feon, V., Schermann-Legionnet, A., Delettre, Y., Aviron, S., Billeter, R., Bugter, R.,

Hendrickx, F., and Burel, F. 2010. Intensification of agriculture, landscape composition and wild bee communities: a large scale study in four European countries. Agriculture, Ecosystems and Environment, 137: 143–150. doi: 10.1016/j.agee.2010.01.015.

Folk, R.L. 1951. A comparison chart for visual percentage estimation. Journal of Sedimentary

Petrology, 21: 32–33.

Fournier, D.A., Skaug, H.J., Ancheta, J., Ianelli, J., Magnusson, A., Maunder, M.N., Nielsen, A., and Sibert, J. 2012. AD Model Builder: using automatic differentiation for statistical inference of highly parameterized complex nonlinear models. Optimization Methods and

Software, 27: 233–249. doi: 10.1080/10556788.2011.597854.

Galbraith, S.M., Cane, J.H., Moldenke, A.R., and Rivers, J.W. 2019. Wild bee diversity increases with local fire severity in a fire-prone landscape. Ecosphere, 10: e02668. doi: 10.1002/ecs2.2668.

Gathmann, A., and Tscharntke, T. 2002. Foraging ranges of solitary bees. Journal of Animal Ecology, 71: 757–764.

Gerla, P.J., Cornett, M.W., Ekstein, J.D., and Ahlering, M.A. 2012. Talking big: Lessons learned from a 9000 hectare restoration in the northern tallgrass prairie. Sustainability, 4: 3066–

3087. doi: 10.3390/su4113066.

Geroff, R.K., Gibbs, J., and McCravy, K.W. 2014. Assessing bee (Hymenoptera: Apoidea) diversity of an Illinois restored tallgrass prairie: methodology and conservation considerations. Journal of Insect Conservation, 18: 951–964. doi: 10.1007/s10841-014-

9703-z.

Gibbs, J. 2011. Revision of the metallic Lasioglossum (Dialictus) of eastern North America (Hymenoptera: Halictidae: Halictini). In Zootaxa. doi: 10.11646/%x.

64

Gibbs, J., Ascher, J.S., Rightmyer, M.G., and Isaacs, R. 2017. The bees of Michigan (Hymenoptera: Apoidea: Anthophila), with notes on distribution, taxonomy, pollination, and natural history. Zootaxa, 4352: 1–160.

Gibbs, J., Hanuschuk, E.J., and Shukla-Bergen, S. 2021. Rediscovery of the rare bee Epeoloides pilosulus in Manitoba. Journal of the Kansas Entomological Society, 93: 1–7.

Goulson, D., Nicholls, E., Botías, C., and Rotheray, E.L. 2015. Bee declines driven by combined stress from parasites, pesticides, and lack of flowers. Science, 347: 1255957. doi:

10.1126/science.1255957.

Griffin, S.R., Bruninga-Socolar, B., and Gibbs, J. 2021. Bee communities in restored prairies are structured by landscape and management, not local floral resources. Basic and Applied Ecology, In Press: 1–8.

Griffin, S.R., Bruninga-Socolar, B., Kerr, M.A., Gibbs, J., and Winfree, R. 2017. Wild bee community change over a 26-year chronosequence of restored tallgrass prairie. Restoration Ecology, 25: 650–660. doi: 10.1111/rec.12481.

Griswold, T., Ikerd, H., Droege, S., Pascarella, J.B., and Pickering, J. 2018. Draft guide to the

female Osmia of eastern North America. Available from https://www.discoverlife.org/mp/20q?guide=Osmia_female.

Harrison, T., Gibbs, J., and Winfree, R. 2017. Anthropogenic landscapes support fewer rare bee species. Landscape Ecology, 34: 967–978. doi: 10.1007/s10980-017-0592-x.

Hartig, F., and Lohse, L. 2020. DHARMa: Residual diagnostics for hierarchical (multi-level/mixed) regression models.

Hickman, K.R., Hartnett, D.C., Cochran, R.C., and Owensby, C.E. 2004. Grazing management effects on plant species diversity in tallgrass prairie. Journal of Range Management, 57: 58–

65.

Hill, M.O. 1973. Diversity and evenness: a unifying notation and its consequences. Ecology, 54: 427–432.

Hines, H.M., and Hendrix, S.D. 2005. Bumble bee (Hymenoptera: Apidae) diversity and

abundance in tallgrass prairie patches: effects of local and landscape floral resources. Environmental Entomology, 34: 1477–1484.

Holzschuh, A., Steffan-dewenter, I., and Tscharntke, T. 2010. How do landscape composition and configuration, organic farming and fallow strips affect the diversity of bees, wasps and

their parasitoids? Journal of Animal Ecology, 79: 491–500. doi: 10.1111/j.1365-2656.2009.01642.x.

Hopfenmuller, S., Steffan-Dewenter, I., and Holzschuh, A. 2014. Trait-specific responses of wild bee communities to landscape composition, configuration and local factors. PLoS ONE, 9:

1–10. doi: 10.1371/journal.pone.0104439.

Hopwood, J.L. 2008. The contribution of roadside grassland restorations to native bee conservation. Biological Conservation, 141: 2632–2640. doi: 10.1016/j.biocon.2008.07.026.

65

Howe, H.F. 1994. Managing species diversity in tallgrass prairie: assumptions and implications. Conservation Biology, 8: 691–704.

Howe, H.F. 2011. Fire season and prairie forb richness in a 21-y experiment. Ecoscience, 18:

317–328. doi: 10.2980/18-4-3421.

Hsieh, T.C., Ma, K.H., and Chao, A. 2019. iNEXT: Interpolation and extrapolation for species diversity. Available from http://chao.stat.nthu.edu.tw/blog/software-download/.

Hudewenz, A., Klein, A.-M., Scherber, C., Stanke, L., Tscharntke, T., Vogel, A., Weigelt, A.,

Weisser, W.W., and Ebeling, A. 2012. Herbivore and pollinator responses to grassland management intensity along experimental changes in plant species richness. Biological Conservation, 150: 42–52. doi: 10.1016/j.biocon.2012.02.024.

Jacobson, M.M., Tucker, E.M., Mathiasson, M.E., and Rehan, S.M. 2018. Decline of bumble

bees in northeastern North America, with special focus on Bombus terricola. Biological Conservation, 217: 437–445. doi: 10.1016/j.biocon.2017.11.026.

Jauker, B., Krauss, J., Jauker, F., and Steffan-Dewenter, I. 2013. Linking life history traits to pollinator loss in fragmented calcareous grasslands. Landscape Ecology, 28: 107–120. doi:

10.1007/s10980-012-9820-6.

Joshi, N.K., Leslie, T., Rajotte, E.G., Kammerer, M.A., Otieno, M., and Biddinger, D.J. 2015. Comparative trapping efficiency to characterize bee abundance, diversity, and community composition in apple orchards. Annals of the Entomological Society of America, 108: 785–

799. doi: 10.1093/aesa/sav057.

Jost, L. 2007. Partitioning diversity into independent alpha and beta components. Ecology, 88: 2427–2439.

Joyce, J., and Morgan, J.P. 1989. Manitoba’s tall-grass prairie conservation project. Proceedings

of the North American Prairie Conferences, 34: 71–74.

Kelleher, E.M., and Choi, Y.D. 2020. Role of plant diversity on arthropod communities in a restored tallgrass prairie of the U.S. Midwest. Restoration Ecology, 28: 1464–1474. doi: 10.1111/rec.13238.

Kennedy, C.M., Lonsdorf, E., Neel, M.C., Williams, N.M., Taylor, H., Winfree, R., Brittain, C., Alana, L., and Cariveau, D. 2013. A global quantitative synthesis of local and landscape effects on wild bee pollinators in agroecosystems. Ecology Letters, 16: 584–599. doi: 10.1111/ele.12082.

van Klink, R., van der Plas, F., (Toos) van Noordwijk, C.G.E., WallisDeVries, M.F., and Olff, H. 2015. Effects of large herbivores on grassland arthropod diversity. Biological Reviews, 90: 347–366. doi: 10.1111/brv.12113.

Knapp, A.K., Blair, J.M., Briggs, J.M., Collins, S.L., Hartnett, D.C., Johnson, L.C., and Towne,

E.G. 1999. Keystone role of Bison in American tallgrass prairie: Bison increase habitat heterogeneity and alter a broad array of plant, community, and ecosystem processes. BioScience, 49: 39–50.

Kohler, M., Sturm, A., Sheffield, C.S., Carlyle, C.N., and Manson, J.S. 2020. Native bee

66

communities vary across three prairie ecoregions due to land use, climate, sampling method and bee life history traits. Insect Conservation and Diversity, 13: 571–584. doi: 10.1111/icad.12427.

Koper, N., Mozel, K.E., and Henderson, D.C. 2010. Recent declines in northern tall-grass prairies and effects of patch structure on community persistence. Biological Conservation, 143: 220–229. doi: 10.1016/j.biocon.2009.10.006.

Kormann, U., Rösch, V., Batáry, P., Tscharntke, T., Orci, K.M., Samu, F., and Scherber, C.

2015. Local and landscape management drive trait-mediated biodiversity of nine taxa on small grassland fragments. Diversity and Distributions, 21: 1204–1217. doi: 10.1111/ddi.12324.

Kwaiser, K.S., and Hendrix, S.D. 2008. Diversity and abundance of bees (Hymenoptera:

Apiformes) in native and ruderal grasslands of agriculturally dominated landscapes. Agriculture, Ecosystems and Environment, 124: 200–204. doi: 10.1016/j.agee.2007.09.012.

LaBerge, W.E. 1956a. A revision of the genus Melissodes in North and Central America Part I (Hymenoptera, Apidae). The University of Kansas Science Bulletin, 37: 911–1194.

LaBerge, W.E. 1956b. A revision of the genus Melissodes in North and Central America Part II (Hymenoptera, Apidae). The University of Kansas Science Bulletin, 38: 533–578.

LaBerge, W.E. 1961. A revision of the bees of the genus Melissodes in North and Central America Part III (Hymenoptera, Apidae). The University of Kansas Science Bulletin, 42:

283–663.

Lane, I.G., Herron-sweet, C.R., Portman, Z.M., and Cariveau, D.P. 2020. Floral resource diversity drives bee community diversity in prairie restorations along an agricultural landscape gradient. Journal of Applied Ecology, In Press: 1–30. doi: 10.1111/1365-

2664.13694.

Larson, D.L., Hernández, D.L., Larson, J.L., Leone, J.B., and Pennarola, N. 2020. Management of remnant tallgrass prairie by grazing or fire: effects on plant communities and soil properties. Ecosphere, 11: e03213. doi: 10.1002/ecs2.3213.

Laverty, T.M., and Harder, L.D. 1988. The bumble bees of eastern Canada. The Canadian Entomologist, 120: 965–987.

Le, S., Josse, J., and Husson, F. 2008. FactoMineR: an R package for multivariate analysis. Journal of Statistical Software, 25: 1–18.

Legendre, P., and Anderson, M.J. 1999. Distance-based redundancy analysis : Testing multispecies responses in ... Ecological Monographs, 69: 1–24.

Leong, J.M., and Thorp, R.W. 1999. Colour-coded sampling: The pan trap colour preferences of oligolectic and nonoligolectic bees associated with a vernal pool plant. Ecological

Entomology, 24: 329–335. doi: 10.1046/j.1365-2311.1999.00196.x.

Ludecke, D., Makowski, D., Waggoner, P., and Patil, I. 2020. Assesment of regression models. Available from https://easystats.github.io/performance/.

67

Manitoba Conservation and Water Stewardship. 2017. 2017 Water availability and drought conditions report.

Manitoba Conservation and Water Stewardship. 2018. 2018 Water availability and drought

conditions report.

Martinez Arbizu, P. 2017. Pairwise multilevel comparison using Adonis.

Mathiasson, M.E., and Rehan, S.M. 2019. Status changes in the wild bees of north -eastern North America over 125 years revealed through museum specimens. Insect Conservation and

Diversity, 12: 278–288. doi: 10.1111/icad.12347.

McCravy, K.W., Geroff, R.K., and Gibbs, J. 2019. Bee (Hymenoptera: Apoidea: Anthophila) functional traits in relation to sampling methodology in a restored tallgrass prairie. Florida Entomologist, 102: 134–140.

Mola, J.M., and Williams, N.M. 2018. Fire-induced change in floral abundance, density, and phenology benefits bumble bee foragers. Ecosphere, 9: 1–9. doi: 10.1002/ecs2.2056.

Moran, M.D. 2014. Bison grazing increases arthropod abundance and diversity in a tallgrass prairie. Environmental Entomology, 43: 1174–1184.

Morandin, L.A., Winston, M.L., Abbott, V.A., and Franklin, M.T. 2007. Can pastureland increase wild bee abundance in agriculturally intense areas? Basic and Applied Ecology, 8: 117–124. doi: 10.1016/j.baae.2006.06.003.

Neary, D.G., and Leonard, J.M. 2020. Effects of fire on grassland soils and water: a review. In

Grasses and Grassland Aspects. Edited by V.M. Kindomihou. IntechOpen, p.

Neumüller, U., Burger, H., Krausch, S., Blüthgen, N., and Ayasse, M. 2020. Interactions of local habitat type, landscape composition and flower availability moderate wild bee communities. Landscape Ecology, 35: 2209–2224. doi: 10.1007/s10980-020-01096-4.

New, T.R. 2014. Insects, fire and conservation. Springer International Publishing, Cham, Switzerland. p. doi: 10.1007/978-3-319-08096-3.

Odanaka, K.A., and Rehan, S.M. 2020. Wild bee distribution near forested landscapes is dependent on successional state. Forest Ecosystems, 7: 1–13. doi:

https://doi.org/10.1186/s40663-020-00241-4.

Oksanen, J., Blanchet, F.G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P., O’Hara, R.B., Simpson, G.L., Solymos, P., Stevens, M.H.H., Szoecs, E., and Wagner, H. 2018. vegan: community ecology package. Available from https://cran.r-

project.org/package=vegan.

Olynyk, M. 2017. Effects of habitat loss, fragmentation, and alteration on wild bees and pollination services in fragmented Manitoba grasslands. [degree] thesis, University of Manitoba, .

Olynyk, M., Westwood, A.R., and Koper, N. 2021. Effects of natural habitat loss and edge effects on wild bees and pollination services in remnant prairies. Environmental Entomology, In Press: 1–12. doi: 10.1093/ee/nvaa186.

68

Orford, K.A., Vaughan, I.P., and Memmott, J. 2015. The forgotten flies: the importance of non-syrphid Diptera as pollinators. Proceedings of the Royal Society B: Biological Sciences, 282: 20142934. doi: http://dx.doi.org/10.1098/rspb.2014.2934.

Packer, L., Genaro, J.A., and Sheffield, C.S. 2007. The bee genera of eastern Canada. Canadian Journal of Arthropod Identification,: 1–32. doi: 10.3752/cjai.2007.03.

Panzer, R. 2002. Compatibility of prescribed burning with the conservation of insects in small, isolated prairie reserves. Conservation Biology, 16: 1296–1307.

Patenaude, A. 2007. Diversity, composition and seasonality of wild bees (Hymenoptera: Apoidea) in a northern mixed-grass prairie preserve. [degree] thesis, University of Manitoba, .

Pennarola, P. 2019. The revery alone won’t do: fire, grazing, and other drivers of bee

communities in remnant tallgrass prairie. [degree] thesis, The University of Minnesota, .

Potts, S.G., Biesmeijer, J.C., Kremen, C., Neumann, P., Schweiger, O., and Kunin, W.E. 2010. Global pollinator declines: trends, impacts and drivers. Trends in Ecology & Evolution, 25: 345–353. doi: 10.1016/j.tree.2010.01.007.

Prendergast, K.S., Menz, M.H.M., Dixon, K.W., and Bateman, P.W. 2020. The relative performance of sampling methods for native bees: an empirical test and review of the literature. Ecosphere, 11: e03076. doi: 10.1002/ecs2.3076.

Purvis, E.E.N., Vickruck, J.L., Best, L.R., Devries, J.H., and Galpern, P. 2020. Wild bee

community recovery in restored grassland-wetland complexes of prairie North America. Biological Conservation, 252: 108829. doi: 10.1016/j.biocon.2020.108829.

Pyne, S.J. 1986. “These conflagrated prairies” a cultural fire history of the grasslands. In The Prairie: past, present, and future: Proceedings of the ninth North American prairie

conference part 6. Fire and prairie ecosystems. Edited by G.K. Clambey and R.H. Pemble.

R Core Team. 2017. R: a language and environment for statistical computing. Available from https://www.r-project.org.

Rényi, A. 1961. On measures of entropy and information. In 4th Berkeley symposium on

mathematical statistics and probability. Edited by J. Neyman.

Rhoades, P., Griswold, T., Waits, L., Bosque-Pérez, N.A., Kennedy, C.M., and Eigenbrode, S.D. 2017. Sampling technique affects detection of habitat factors influencing wild bee communities. Journal of Insect Conservation, 21: 703–714. doi: 10.1007/s10841-017-0013-

0.

Rivers-Moore, J., Andrieu, E., Vialatte, A., and Ouin, A. 2020. Wooded semi-natural habitats complement permanent grasslands in supporting wild bee diversity in agricultural landscapes. Insects, 11: 1–21. doi: 10.3390/insects11110812.

Robson, D.B. 2008. The structure of the flower–insect visitor system in tall-grass prairie. Botany, 86: 1266–1278. doi: 10.1139/B08-083.

Robson, D.B., Hamel, C., Neufeld, R., and Bleho, B.I. 2019. Habitat filtering influences plant–

69

pollinator interactions in prairie ecosystems. Botany, 97: 204–220.

Rosenberger, D.W., and Conforti, M.L. 2020. Native and agricultural grassland use by stable and declining bumble bees in Midwestern North America. Insect Conservation and Diversity,.

doi: 10.1111/icad.12448.

Rotchés-Ribalta, R., Winsa, M., Roberts, S.P.M., and Öckinger, E. 2018. Associations between plant and pollinator communities under grassland restoration respond mainly to landscape connectivity. Journal of Applied Ecology, 55: 2822–2833. doi: 10.1111/1365-2664.13232.

Roulston, T.H., and Goodell, K. 2011. The role of resources and risks in regulating wild bee populations. Annual Review of Entomology, 56: 293–312. doi: 10.1146/annurev-ento-120709-144802.

Roulston, T.H., Smith, S.A., and Brewster, A.L. 2007. A comparison of pan trap and intensive

net sampling techniques for documenting a bee (Hymenoptera: Apiformes) fauna. Journal of the Kansas Entomological Society, 80: 179–181.

Samson, F., and Knopf, F. 1994. Prairie conservation in North America. BioScience, 44: 418–421.

Sardiñas, H.S., and Kremen, C. 2014. Evaluating nesting microhabitat for ground-nesting bees using emergence traps. Basic and Applied Ecology, 15: 161–168. doi: 10.1016/j.baae.2014.02.004.

Schindler, M., Diestelhorst, O., Härtel, S., Saure, C., Schanowski, A., and Schwenninger, H.R.

2013. Monitoring agricultural ecosystems by using wild bees as environmental indicators. BioRisk, 71: 53–71. doi: 10.3897/biorisk.8.3600.

Semmler, S.J. 2015. Community composition and pollination network structure in a fire managed Canadian tall grass prairie. [degree] thesis, University of Manitoba, .

Shaw, R.F., Phillips, B.B., Doyle, T., Pell, J.K., Redhead, J.W., Savage, J., Woodcock, B.A., Bullock, J.M., and Osborne, J.L. 2020. Mass-flowering crops have a greater impact than semi- natural habitat on crop pollinators and pollen deposition. Landscape Ecology, 35: 513–527. doi: 10.1007/s10980-019-00962-0.

Sheffield, C.S., Danae Frier, S., and Dumesh, S. 2014. The bees (Hymenoptera: Apoidea, Apiformes) of the prairies ecozone with comparisons to other grasslands of Canada. In Arthropods of Canadian Grasslands (Volume 4): Biodiversity and Systematics Part 2. Edited by D.J. Giberson and H.A. Carcamo. Biological Survey of Canada, pp. 427–467.

Shorthouse, D.P. 2010. SimpleMappr, an online tool to produce publication-quality point maps.

Sjödin, N.E., Bengtsson, J., and Ekbom, B. 2008. The influence of grazing intensity and landscape composition on the diversity and abundance of flower-visiting insects. Journal of Applied Ecology, 45: 763–772. doi: 10.1111/j.1365-2664.2007.01443.x.

Skaug, H., Fournier, D., Nielsen, A., Magnusson, A., and Bolker, B. 2012. glmmADMB. Available from http://www.admb-project.org/.

Smith, G.W., Debinski, D.M., Scavo, N.A., Lange, C.J., John, T., Moranz, R.A., Miller, J.R.,

70

Engle, D.M., and Toth, A.L. 2016. Bee abundance and nutritional status in relation to grassland management practices in an agricultural landscape. Environmental Entomology, 45: 338–347. doi: 10.1093/ee/nvw005.

Spiesman, B.J., Bennett, A., Isaacs, R., and Gratton, C. 2019. Harvesting effects on wild bee communities in bioenergy grasslands depend on nesting guild. Ecological Applications, 29: e01828. doi: 10.1002/eap.1828.

Steckel, J., Westphal, C., Peters, M.K., Bellach, M., Rothenwoehrer, C., Erasmi, S., Scherber, C.,

Tscharntke, T., and Steffan-Dewenter, I. 2014. Landscape composition and configuration differently affect trap-nesting bees, wasps and their antagonists. Biological Conservation, 172: 56–64. doi: 10.1016/j.biocon.2014.02.015.

Stein, D.S., Debinski, D.M., Pleasants, J.M., and Toth, A.L. 2020. Evaluating native bee

communities and nutrition in managed grasslands. Environmental Entomology, 49: 717–725. doi: 10.1093/ee/nvaa009.

Stephen, W.P., and Rao, S. 2005. Unscented color traps for non-Apis bees (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society, 78: 373–380.

Sveinson, J.M.A. 2003. Restoring tallgrass prairie in southern Manitoba, Canada. [degree] thesis, University of Manitoba, .

Taft, J.B., and Kron, Z.P. 2014. Evidence of species and functional group attrition in shrub-encroached prairie: implications for restoration. American Midland Naturalist, 172: 252–

265.

Tonietto, R.K., Ascher, J.S., and Larkin, D.J. 2017. Bee communities along a prairie restoration chronosequence: similar abundance and diversity, distinct composition. Ecological Applications, 27: 705–717.

Towne, E.G., Hartnett, D.C., and Cochran, R.C. 2005. Vegetation trends in tallgrass prairie from bison and cattle grazing. Ecological Applications, 15: 1550–1559.

Towne, E.G., and Kemp, K.E. 2003. Vegetation dynamics from annually burning tallgrass prairie in different seasons. Journal of Range Management, 56: 185–192.

Trimble, S.W., and Mendel, A.C. 1995. The cow as a geomorphic agent - a critical review. Geomorphology, 13: 233–253.

Twidwell, D., West, A.S., Hiatt, W.B., Ramirez, A.L., Winter, J.T., Engle, D.M., Fuhlendorf, S.D., and Carlson, J.D. 2016. Plant invasions or fire policy: which has altered fire behavior

more in tallgrass prairie? Ecosystems, 19: 356–368. doi: 10.1007/s10021-015-9937-y.

Veech, J.A. 2006. A comparison of landscapes occupied by increasing and decreasing populations of grassland birds. Conservation Biology, 20: 1422–1432. doi: 10.1111/j.1523-1739.2006.00487.x.

Veen, G.F.C., Blair, J.M., Smith, M.D., and Collins, S.L. 2008. Influence of grazing and fire frequency on small-scale plant community structure and resource variability in native tallgrass prairie. Oikos, 117: 859–866. doi: 10.1111/j.2008.0030-1299.16515.x.

71

Vickruck, J.L., Best, L.R., Gavin, M.P., Devries, J.H., and Galpern, P. 2019. Pothole wetlands provide reservoir habitat for native bees in prairie croplands. Biological Conservation, 232: 43–50. doi: 10.1016/j.biocon.2019.01.015.

Wagle, P., and Gowda, P.H. 2018. Tallgrass prairie responses to management practices and disturbances: a review. Agronomy, 8: 1–16. doi: 10.3390/agronomy8120300.

Walters, C.M., and Martin, M.C. 2003. An examination of the effects of grazing on vegetative and soil parameters in the tallgrass prairie. Transactions of the Kansas Academy of Science,

106: 59–70.

Weiner, C.N., Werner, M., Linsenmair, K.E., and Blüthgen, N. 2011. Land use intensity in grasslands: changes in biodiversity, species composition and specialisation in flower visitor networks. Basic and Applied Ecology, 12: 292–299. doi: 10.1016/j.baae.2010.08.006.

Whiles, M.R., and Charlton, R.E. 2006. The ecological significance of tallgrass prairie arthropods. Annual Review of Entomology, 51: 387–412. doi: 10.1146/annurev.ento.51.110104.151136.

Williams, N.M., Crone, E.E., Roulston, T.H., Minckley, R.L., Packer, L., and Potts, S.G. 2010.

Ecological and life-history traits predict bee species responses to environmental disturbances. Biological Conservation, 143: 2280–2291. doi: 10.1016/j.biocon.2010.03.024.

Wright Morton, L., Regen, E., Engle, D.M., Miller, J.R., and Harr, R.N. 2010. Perceptions of landowners concerning conservation, grazing, fire, and eastern redcedar management in

tallgrass prairie. Rangeland Ecology & Management, 63: 645–654.

Wrobleski, D.W., and Kauffman, J.B. 2003. Initial effects of prescribed fire on morphology, abundance, and phenology of forbs in big sagebrush communities in southeastern Oregon. Restoration Ecology, 11: 82–90.

Zayed, A., and Packer, L. 2005. Complementary sex determination substantially increases extinction proneness of haplodiploid populations. Proceedings of the National Academy of Sciences of the United States of America, 102: 10742–6. doi: 10.1073/pnas.0502271102.

Zurbuchen, A., Landert, L., Klaiber, J., Müller, A., Hein, S., and Dorn, S. 2010. Maximum

foraging ranges in solitary bees: only few individuals have the capability to cover long foraging distances. Biological Conservation, 143: 669–676. doi: 10.1016/j.biocon.2009.12.003.

72

2.8 Chapter 2 Supplementary Material

Figure 2.11 Graph showing the first two axes of a PCA performed on groundcover data obtained

by averaging two rounds of quadrat sampling performed in 2018. 16 quadrats were recorded

from each study site and then averaged by site. This was done twice, once in June, and once in

August. The site averages of the two rounds were once again averaged to obtain the results put in

to the PCA.

73

Figure 2.12 Graph of the first two axes of a principal coordinate analysis performed on percent cover on the various landscapes found at 1 km from the centre of study sites.

74

Bee Model Tables

Table 2.2 Results of GLMMs performed on bee bowl capture abundance, species richness, and Hill’s Shannon Effective Species Number. Site was used as a random effect in all models. In total there were 126 observations, with 18 observations of ‘Site’. Landcover variables were

measured using a 1 km radius. The first axis of PCAs for landcover and groundcover were used. The models represent the submodel with the lowest AICc score.

Response Variable

Fixed Effects Est. Std. Error z value p value

Abundance

Intercept 1.905 1.042 1.83 0.067

Year (2017) 0.693 0.168 4.13 <0.001

Day-of-year -0.02 0.002 -8.93 <0.001

Avg. Temp. 0.077 0.035 2.19 0.028

Avg. Precip. -0.14 0.056 -2.55 0.011

Shannon Div. 1.894 0.447 4.23 <0.001

Landcover PCA 0.135 0.044 3.04 0.002

Treatment: Grazed -0.31 0.199 -1.57 0.116

Treatment: Undisturbed 0.071 0.221 0.32 0.747

Groundcover PCA 0.109 0.044 2.48 0.013

Species Richness

Intercept -0.17 0.524 -0.33 0.745

Year (2017) -0.49 0.105 -4.64 <0.001

BV (Yes) -0.70 0.125 -5.64 <0.001

Avg. Precip. -0.10 0.056 -1.86 0.062

Shannon Div. @ 1 km 1.424 0.310 4.59 <0.001

Landcover PCA @ 1 km 0.109 0.032 3.37 <0.001

Treatment: Grazed -0.28 0.137 -2.07 0.038

Treatment: Undisturbed 0.108 0.147 0.73 0.463

Hill’s Shannon

Effective Species Number

Intercept -0.67 0.576 -1.16 0.246

Year (2017) -0.34 0.117 -2.89 0.004

BV (Yes) -0.49 0.128 -3.81 <0.001

Shannon Div. @ 1 km 1.243 0.307 4.04 <0.001

Landcover PCA @ 1 km 0.088 0.031 2.89 0.004

Landscape MSI @ 1 km 0.156 0.10 1.57 0.116

Treatment: Grazed -0.23 0.139 -1.65 0.098

Treatment: Undisturbed 0.10 0.156 0.64 0.522

Model Distribution Dispersion Parameter

Random Effect S.D.

AIC Log

Likelihood R2C

75

Abundance Negative Binomial

2.52 (+/- 0.43) 0.103 827.2 -401.588 0.219

Sp. Rich. Qausi-

Poisson 1.42 (+/- 0.20) 0.061 612.2 -296.115 0.672

Hill’s Shannon

Gamma (Log Link)

3.43 (+/- 0.41) <0.001 551.9 -265.948 0.273

Table 2.3 Results of GLMMs performed on ground-nesting bee bowl capture abundance, species richness, and Hill’s Shannon Effective Species Number. Site was used as a random effect in all

models. The models represent the submodel with the lowest AICc score.

Response

Variable Fixed Effects Est. Std. Error z value p value

Abundance

Intercept 1.397 1.208 1.16 0.248

Year (2017) 0.722 0.180 4.00 <0.001

Day-of-year -0.02 0.003 -7.41 <0.001

Avg. Temp. 0.083 0.039 2.14 0.032

Avg. Precip. -0.14 0.061 -2.28 0.023

Shannon Div. @ 1 km 1.646 0.552 2.98 0.003

Landcover PCA @ 1 km 0.154 0.045 3.42 0.001

Groundcover PCA 0.121 0.056 2.16 0.030

Species

Richness

Intercept -0.89 0.572 -1.55 0.120

Year (2017) -0.19 0.119 -1.60 0.109

BV (Yes) -0.36 0.137 -2.66 0.008

Avg. Precip. -0.12 0.059 -1.95 0.051

Shannon Div. @ 1 km 1.517 0.339 4.47 <0.001

Landcover PCA @ 1 km 0.130 0.036 3.59 <0.001

Treatment: Grazed -0.25 0.150 -1.68 0.094

Treatment: Undisturbed 0.114 0.163 0.70 0.485

Groundcover PCA 0.034 0.033 1.02 0.307

Intercept -2.19 0.765 -2.87 0.004

76

Hill’s

Shannon

Effective

Species

Number

Year (2017) -0.57 0.251 -2.29 0.022

Day-of-year 0.007 0.003 2.43 0.015

BV (Yes) -0.63 0.218 -2.89 0.004

Shannon Div. @ 1 km 1.023 0.330 3.10 0.002

Landcover PCA @ 1 km 0.105 0.028 3.71 <0.001

Landscape MSI @ 1 km 0.243 0.109 2.23 0.026

Model Distribution Dispersion

Parameter

Random

Effect S.D. AIC

Log

Likelihood R2C

Abundance Negative

Binomial 2.20 (+/- 0.39) 0.239 786.2 -383.114 0.231

Sp. Rich. Qausi-

Poisson 1.22 (+/- 0.19) 0.019 545.2 -261.582 0.440

Hill’s Shannon

Gamma (Log Link)

3.65 (+/- 0.47) 0.131 469.9 -225.971 0.265

Table 2.4 Results of GLMMs performed on stem-nesting bee bowl capture abundance, species richness, and Hill’s Shannon Effective Species Number. Site was used as a random effect in all models. The models represent the submodel with the lowest AICc score.

Response

Variable Fixed Effects Est. Std. Error z value p value

Abundance

Intercept 2.680 1.283 2.09 0.037

Day-of-year -0.02 0.003 -7.45 <0.001

BV (Yes) -0.51 0.271 -1.87 0.062

Shannon Div. @ 1 km 1.761 0.725 2.43 0.015

Treatment: Grazed -0.80 0.312 -2.55 0.011

Treatment: Undisturbed -0.29 0.288 -0.99 0.323

Species

Richness

Intercept 1.927 0.913 2.11 0.035

Day-of-year -0.02 0.002 -6.29 <0.001

BV (Yes) -0.46 0.265 -1.73 0.083

77

Groundcover 0.082 0.056 1.46 0.143

Treatment: Grazed -0.51 0.230 -2.23 0.026

Treatment: Undisturbed -0.17 0.206 -0.84 0.399

Hill’s

Shannon

Effective

Species

Number

Intercept 2.433 0.237 10.27 <0.001

Day-of-year -0.01 0.001 -8.21 <0.001

Avg. Precip. -0.10 0.037 -2.62 0.009

Model Distribution Dispersion Parameter

Random Effect S.D.

AIC Log

Likelihood R2C

Abundance Negative Binomial

1.62 (+/- 0.43) 0.239 464.2 -224.076 0.432

Sp. Rich. Quasi-

Poisson 1.50 (+/-0.22) 0.015 398.8 -190.395 0.671

Hill’s

Shannon

Gamma

(Log Link) 4.89 (+/- 0.64) 0.197 294.3 -142.132 0.471

Table 2.5 Results of GLMMs performed on blue vane bee capture abundance, species richness, and Hill’s Shannon Effective Species Number. Site was used as a random effect in all models. The models represent the submodel with the lowest AICc score.

Response

Variable Fixed Effects Est. Std. Error z value p value

Abundance

Intercept -7.92 1.401 -5.65 <0.001

Day-of-year 0.040 0.006 6.29 <0.001

Avg. Temp. 0.163 0.058 2.82 0.005

Avg. Precip. 0.287 0.077 3.75 <0.001

Intercept 18.287 4.662 3.92 <0.001

78

Species

Richness

Day-of-year -0.05 0.021 -2.22 0.027

Avg. Precip. 0.735 0.238 3.08 0.002

Hill’s

Shannon

Effective

Species

Number

Intercept 8.287 0.687 12.07 <0.001

Day-of-year -0.02 0.002 -9.17 <0.001

Avg. Temp. -0.14 0.024 -5.79 <0.001

Landcover PCA @ 1 km 0.049 0.019 2.56 0.010

Model Distribution

Dispersion

Parameter/Residual Variance

Random

Effect S.D.

AIC Log

Likelihood R2C

Abundance Negative Binomial

5.41 (+/- 1.42) 0.001 328.9 -158.447 0.109

Sp. Rich. Gaussian 1.63 (+/- 0.26) 0.038 147.6 -68.7832 0.501

Hill’s Shannon

Gamma (Log Link)

27.07 (+/- 8.52) 0.105 113.9 -50.9339 0.793

79

Table 2.6 List of species caught in bee bowls in 2017 along with their nesting guild (G = ground,

S = stem (above ground) and distribution of numbers caught between the three treatments: B =

burned, G = grazed, U = undisturbed. The other columns represent the types of bee bowls they

were caught in, with B.G. = blue ground bowl, Y.G. = yellow ground, W.G. = white ground,

B.E. = blue elevated, Y.E. = yellow elevated, and W.E. = white elevated.

Species Author Nest B G U B.G. Y.G. W.G.

B.E. Y.E. W.E. Total

Agapostemon

texanus

Cresson 1872 G 0 0 1 1 0 0 0 0 0 1

Agapostemon

virescens

(Fabricius 1775) G 0 0 1 0 0 0 1 0 0 1

Anthophora

terminalis

Cresson 1869 S 6 3 7 8 0 2 5 1 0 16

Augochlorella aurata (Smith 1853) G 3 1 4 4 2 2 0 0 0 8

Bombus borealis Kirby 1837 G 9 6 3 11 1 0 6 0 0 18

Bombus fervidus (Fabricius 1798) G 6 3 8 4 0 0 4 6 3 17

Bombus griseocollis (De Geer 1773) G 0 0 1 0 0 0 1 0 0 1

Bombus rufocinctus Cresson, 1863 G 5 1 5 4 0 1 5 1 0 11

Bombus sandersoni Franklin 1913 G 1 0 2 0 0 1 1 0 1 3

Bombus ternarius Say 1837 G 2 2 3 1 0 1 4 1 0 7

Bombus vagans Smith 1854 G 7 8 11 4 0 0 15 4 3 26

Ceratina mikmaqi Rehan & Sheffield

2011

S 0 0 1 1 0 0 0 0 0 1

Coelioxys porterae Cockerell 1900 S 0 0 1 0 0 0 0 0 1 1

Dianthidium pudicum (Cresson 1879) Surf. 0 1 0 1 0 0 0 0 0 1

Halictus confusus Smith 1853 G 0 4 2 3 3 0 0 0 0 6

Halictus rubicundus (Christ 1791) G 3 0 4 0 2 2 0 1 2 7

Heriades carinata Cresson 1864 S 1 1 0 0 0 0 0 2 0 2

Heriades variolosa (Cresson 1872) S 1 1 2 0 0 1 1 0 2 4

Hoplitis pilosifrons (Cresson 1864) S 5 4 2 4 1 5 0 1 0 11

Hoplitis producta (Cresson 1864) S 1 0 2 0 1 2 0 0 0 3

Hoplitis spoliata (Provancher 1888) S 1 1 5 0 4 3 0 0 0 7

Hylaeus affinis (Smith 1853) S 0 0 2 1 1 0 0 0 0 2

Hylaeus annulatus (Linnaeus 1758) S 0 0 1 0 0 0 1 0 0 1

Lasioglossum admirandum

(Sandhouse 1924) G 19 13 17 17 1 11 13 0 7 49

Lasioglossum albipenne

(Robertson 1890) G 0 0 1 0 0 0 1 0 0 1

Lasioglossum cinctipes

(Provancher 1888) G 0 0 1 0 0 0 0 1 0 1

Lasioglossum leucocomus

(Lovell 1908) G 0 1 0 0 0 0 1 0 0 1

Lasioglossum leucozonium

(Schrank 1781) G 2 1 7 1 0 0 3 2 4 10

Lasioglossum nigroviride

(Graenicher 1910) S 0 1 0 1 0 0 0 0 0 1

Lasioglossum novascotiae

(Mitchell 1960) G 1 0 0 0 0 0 0 0 1 1

Lasioglossum paraforbesii

McGinley 1986 G 1 0 0 0 0 1 0 0 0 1

Lasioglossum pilosum (Smith 1853) G 0 0 1 1 0 0 0 0 0 1

Lasioglossum planatum

(Lovell 1905) G 4 18 5 12 3 7 2 1 2 27

Lasioglossum sagax (Sandhouse 1924) G 1 0 1 0 1 1 0 0 0 2

Lasioglossum semicaeruleum

(Cockerell 1895) G 2 1 0 2 0 1 0 0 0 3

Lasioglossum viridatum

Lovell 1905 G 58 90 104 124 4 30 60 5 29 252

Lasioglossum zonulum

(Smith 1848) G 5 10 16 11 1 8 6 3 2 31

Megachile gemula Cresson 1878 S 0 0 1 0 0 0 0 1 0 1

Megachile inermis Provancher 1888 S 1 0 0 0 0 0 1 0 0 1

Megachile latimanus Say 1823 G 1 0 0 1 0 0 0 0 0 1

Megachile relativa Cresson 1878 S 0 1 0 0 0 0 1 0 0 1

Melissodes agilis Cresson 1878 G 2 2 5 4 0 0 4 0 1 9

Melissodes druriellus (Kirby 1802) G 0 0 2 0 0 0 2 0 0 2

Melissodes illatus Lovell & Cockerell 1906

G 1 0 2 0 0 1 1 0 1 3

Melissodes rivalis Cresson 1872 G 0 0 1 0 0 0 1 0 0 1

Melissodes trinodis Robertson 1901 G 3 2 5 3 0 0 4 3 0 10

Osmia atriventris Cresson 1864 S 1 0 0 1 0 0 0 0 0 1

Osmia proxima Cresson 1864 S 0 0 1 1 0 0 0 0 0 1

Osmia simillima Smith 1853 S 1 0 1 2 0 0 0 0 0 2

Osmia tersula Cockerell 1912 S 1 0 0 0 0 1 0 0 0 1

Stelis labiata (Provancher 1888) S 1 1 0 0 0 1 0 1 0 2

Total: 51 Species Total

156 177 239 228 25 82 144 34 59 572 Avg. Total per site

31.2

0 29.5

0 34.1

4 374 150 264 110 33 59

Number of Species

32 25 38 602 175 346 254 67 118

80

Avg. Number spp. per site

6.40 4.17 5.43

Table 2.7 List of species caught in bee bowls in 2018 along with their nesting guild (G = ground,

S = stem (above ground), U = unknown) and distribution of numbers caught between the three

disturbance types: B = burned, G = grazed, U = undisturbed. The other columns represent the

types of bee bowls they were caught in, with B.G. = blue ground bowl, Y.G. = yellow ground,

W.G. = white ground, B.E. = blue elevated, Y.E. = yellow elevated, and W.E. = white elevated.

Species Author Nest B G U B.G. Y.G. W.G.

B.E. Y.E. W.E. Total

Agapostemon texanus

Cresson 1872 G 2 0 3 1 0 2 1 0 1 5

Andrena algida Smith 1853 G 1 0 1 0 1 0 1 0 0 2

Andrena commoda Smith 1879 G 1 0 0 1 0 0 0 0 0 1

Andrena cressonii Robertson 1891 G 6 1 1 1 3 4 0 0 0 8

Andrena persimulata Viereck 1917 G 0 1 0 0 1 0 0 0 0 1

Anthophora bomboides

Kirby 1838 G 1 0 1 0 0 0 1 1 0 2

Anthophora

terminalis

Cresson 1869 S 7 7 10 14 3 4 3 0 0 24

Augochlorella aurata (Smith 1853) G 13 5 6 6 7 10 0 1 0 24

Bombus bimaculatus Cresson 1863 G 1 0 1 0 0 0 1 0 1 2

Bombus borealis Kirby 1837 G 4 1 0 3 0 0 1 0 1 5

Bombus fervidus (Fabricius 1798) G 1 3 2 1 0 0 2 0 3 6

Bombus rufocinctus Cresson, 1863 G 2 2 0 1 1 2 0 0 0 4

Bombus sandersoni Franklin 1913 G 1 0 0 0 0 0 0 0 1 1

Bombus ternarius Say 1837 G 1 3 1 1 0 0 3 0 1 5

Bombus vagans Smith 1854 G 7 3 7 4 0 0 7 2 4 17

Ceratina calcarata Robertson 1900 S 0 0 1 0 0 1 0 0 0 1

Ceratina dupla Say 1837 S 1 2 4 1 2 4 0 0 0 7

Ceratina mikmaqi Rehan & Sheffield 2011

S 6 3 8 9 0 8 0 0 0 17

Colletes kincaidii Cockerell 1898 G 0 0 1 0 0 0 1 0 0 1

Dufourea marginata (Cresson 1878) G 0 0 1 0 0 0 1 0 0 1

Halictus confusus Smith 1853 G 4 6 5 3 6 6 0 0 0 15

Halictus ligatus Say 1837 G 1 0 0 0 1 0 0 0 0 1

Halictus rubicundus (Christ 1791) G 9 7 9 8 3 12 1 0 1 25

Heriades carinata Cresson 1864 S 1 0 1 0 0 0 0 1 1 2

Hoplitis pilosifrons (Cresson 1864) S 34 12 15 29 14 17 1 0 0 61

Hoplitis producta (Cresson 1864) S 1 0 2 1 0 2 0 0 0 3

Hoplitis spoliata (Provancher 1888) S 2 1 3 1 2 3 0 0 0 6

Hylaeus affinis (Smith 1853) S 1 1 1 2 1 0 0 0 0 3

Hylaeus annulatus (Linnaeus 1758) S 1 0 0 0 0 0 0 1 0 1

Hylaeus mesillae (Cockerell 1907) S 1 2 1 1 2 1 0 0 0 4

Lasioglossum admirandum

(Sandhouse 1924) G 89 17 23 58 6 29 21 4 11 129

Lasioglossum coriaceum

(Smith 1853) G 2 1 0 2 0 1 0 0 0 3

Lasioglossum inconditum

(Cockerell 1916) G 1 0 1 1 1 0 0 0 0 2

Lasioglossum laevissimum

(Smith 1853) G 0 1 0 0 1 0 0 0 0 1

Lasioglossum leucocomus

(Lovell 1908) G 1 0 1 1 0 0 1 0 0 2

Lasioglossum leucozonium

(Schrank 1781) G 7 5 13 6 9 1 5 0 4 25

Lasioglossum michiganense

(Mitchell 1960) G 1 0 0 0 0 1 0 0 0 1

Lasioglossum pectorale

(Smith 1853) G 1 0 0 0 0 1 0 0 0 1

Lasioglossum perpunctatum

(Ellis 1913) G 2 0 2 0 3 1 0 0 0 4

Lasioglossum planatum

(Lovell 1905) G 4 15 4 11 5 7 0 0 0 23

Lasioglossum sagax (Sandhouse 1924) G 4 11 9 14 2 1 6 1 0 24

Lasioglossum subviridatum

(Cockerell 1938) S 2 0 0 1 0 0 1 0 0 2

Lasioglossum viridatum

Lovell 1905 G 65 96 81 106 16 78 19 11 12 242

Lasioglossum zonulum

(Smith 1848) G 25 31 59 35 19 33 12 6 10 115

Megachile inermis Provancher 1888 S 2 0 3 4 0 0 1 0 0 5

Megachile latimanus Say 1823 G 0 0 1 0 0 0 1 0 0 1

Megachile melanophaea

Smith 1853 G 1 0 0 0 0 0 1 0 0 1

Megachile pugnata Say 1837 S 0 0 1 0 0 0 0 1 0 1

Megachile relativa Cresson 1878 S 0 1 2 0 0 1 0 0 2 3

81

Melissodes agilis Cresson 1878 G 3 6 4 1 0 1 9 1 1 13

Melissodes druriellus (Kirby 1802) G 0 0 1 0 0 0 0 0 1 1

Melissodes trinodis Robertson 1901 G 4 0 3 0 0 2 1 1 3 7

Osmia atriventris Cresson 1864 S 6 15 16 15 11 10 0 1 0 37

Osmia bucephala Cresson 1864 S 5 3 5 4 5 2 2 0 0 13

Osmia lignaria Say 1837 S 0 0 1 0 1 0 0 0 0 1

Osmia proxima Cresson 1864 S 2 5 7 3 5 4 2 0 0 14

Osmia simillima Smith 1853 S 4 5 12 11 5 4 1 0 0 21

Osmia subarcticus Cockerell 1912 U 0 0 3 1 1 1 0 0 0 3

Osmia tersula Cockerell 1912 S 19 7 12 11 13 10 3 0 1 38

Sphecodes pecosensis Cockerell 1904 G 1 0 0 1 0 0 0 0 0 1

Stelis labiata (Provancher 1888) S 0 1 0 0 0 0 0 1 0 1

Total: 61 Species Total

361 280 349 374 150 264 110 33 59 990 Avg. Total per site

72.2

0 46.6

7 49.8

6

Number of Species 49 33 46

Avg. Number of Species per site

9.80 5.50 6.57

82

Table 2.8 List of species caught in blue vane traps in 2018 along with their nesting guild (G =

ground, S = stem (above ground) and distribution of numbers caught between the three

disturbance types: B = burned, G = grazed, U = undisturbed.

Species Author Nest B G U Total

Agapostemon texanus Cresson 1872 G 0 0 1 1

Agapostemon virescens (Fabricius 1775) G 0 1 1 2

Anthophora terminalis Cresson 1869 S 10 22 17 49 Anthophora walshii Cresson 1869 G 1 0 0 1

Augochlorella aurata (Smith 1853) G 0 1 0 1

Bombus bimaculatus Cresson 1863 G 0 0 2 2

Bombus borealis Kirby 1837 G 49 53 70 172 Bombus fervidus (Fabricius 1798) G 22 15 28 65

Bombus griseocollis (De Geer 1773) G 0 0 1 1

Bombus nevadensis Cresson 1874 G 1 2 0 3

Bombus rufocinctus Cresson, 1863 G 6 6 14 26 Bombus ternarius Say 1837 G 7 8 9 24

Bombus terricola Kirby 1837 G 0 2 0 2

Bombus vagans Smith 1854 G 11 6 9 26

Ceratina mikmaqi Rehan & Sheffield 2011 S 0 1 1 2 Colletes kincaidii Cockerell 1898 G 0 0 1 1

Dianthidium pudicum (Cresson 1879) Surf. 0 1 0 1

Halictus parallelus Say 1837 G 0 0 1 1

Halictus rubicundus (Christ 1791) G 6 6 4 16 Heriades variolosa (Cresson 1872) S 1 0 0 1

Hoplitis pilosifrons (Cresson 1864) S 5 2 4 11

Hoplitis spoliata (Provancher 1888) S 0 0 1 1

Hylaeus annulatus (Linnaeus 1758) S 1 3 12 16 Hylaeus mesillae (Cockerell 1907) S 10 1 3 14

Lasioglossum admirandum (Sandhouse 1924) G 1 0 0 1

Lasioglossum laevissimum (Smith 1853) G 0 3 1 4

Lasioglossum leucocomus (Lovell 1908) G 0 2 0 2 Lasioglossum leucozonium (Schrank 1781) G 0 1 1 2

Lasioglossum planatum (Lovell 1905) G 5 1 1 7

Lasioglossum truncatum (Robertson 1901) G 1 0 0 1

Lasioglossum viridatum Lovell 1905 G 20 31 51 102 Lasioglossum zonulum (Smith 1848) G 15 42 167 224

Macropis nuda (Provancher 1882) G 0 0 1 1

Megachile inermis Provancher 1888 S 0 2 2 4

Megachile latimanus Say 1823 G 0 1 0 1 Megachile montivaga Cresson 1878 S 1 0 0 1

Megachile relativa Cresson 1878 S 2 1 1 4

Melissodes agilis Cresson 1878 G 227 331 356 914

Melissodes desponsus Smith 1854 G 1 1 0 2 Melissodes druriellus (Kirby 1802) G 1 1 2 4

Melissodes illatus Lovell & Cockerell 1906 G 2 1 2 5

Melissodes trinodis Robertson 1901 G 82 106 125 313

Osmia atriventris Cresson 1864 S 2 0 0 2 Osmia simillima Smith 1853 S 1 1 4 6

Triepeolus helianthi (Robertson 1897) G 0 0 1 1

Total = 45 species Total 491 655 894 2040

Avg. Total per site 98.2 109.1667 127.71429

Number of Species 27 31 32

Avg. Number of Species per site 5.4 5.166667 4.5714286

83

Table 2.9 List of study sites with corresponding disturbance type and coordinates.

Site Disturbance Type Latitude (° N) Longitude (° W) Stefanchuk 2 Undisturbed 49.134385 -96.740645

Smook SW Undisturbed 49.1677425 -96.74031

Smook NW Undisturbed 49.17568 -96.740368

Townsend Undisturbed 49.163795 -96.67451

Forrester & Casson Undisturbed 49.1185725 -96.66499

Dykun Undisturbed 49.1049425 -96.649898

Clough Undisturbed 49.1576575 -96.649858

Stefanchuk Grazed 49.1553 -96.739845

Nickel Grazed 49.1629125 -96.716645

Fisher Grazed 49.149765 -96.695408

Fisher 2 Grazed 49.1416775 -96.695463

Penner Grazed 49.1205975 -96.654183

Storoschuk Grazed 49.13253 -96.650598

Singh NW Burned 49.148185 -96.74066

Saranchuk Burned 49.179495 -96.67225

Rose Burned 49.171915 -96.672625

Kai Kwong Lee Burned 49.17512 -96.650273

Bott Burned 49.16934 -96.65027

84

Chapter 3: Relationship between the two data chapters (Chapters 2 and 4)

The preceding chapter explored the effects of prairie disturbance and land use on wild

bees; an important group of beneficial insects in the tallgrass prairie. I found that burned and

grazed sites differed in bee community composition, and that grazed sites supported less

abundance, richness, and diversity of bees compared to burned and undisturbed sites. Ground-

nesting bees had less species richness in grazed sites compared to undisturbed sites. Stem-nesting

bees increased in abundance and species richness in burned sites compared to grazed sites. Wild

bee diversity responded positively with increasing amount of forest, ditch, and pasture in the

surrounding landscape. As well, increasing amount of bare ground, forbs, and blooming flower

numbers in the local groundcover positively affected the entire bee community, as well as the

ground-nesting bee community specifically. My results provide insight into how prairies and the

surrounding landscape can be best managed to benefit wild bees.

I performed a parallel study to examine potential trade-offs in management by examining

the effects of prairie disturbance and land use on haplogastran beetles. These beetles provide

different ecological services important to the prairie, especially decomposition and predation of

pest insects. Haplogastran beetles have different habitat requirements to bees and may not

respond in similar fashion to disturbance and land use.

The following chapter focuses on these decomposer and predatory beetles using a similar

experimental design to the preceding bee chapter. By examining the effects of disturbance,

landscape, and landcover on two groups (wild bees and haplogastran beetles) that are essential to

the tallgrass prairie, it is hoped that management policies that favour multiple groups of

beneficial insects can be implemented, while avoiding short sighted decisions that may benefit

one group at the expense of another.

85

Chapter 4: Fire and grazing present trade-offs for an ancestrally detritivorous group of

beetles (Coleoptera: Haplogastra) in tallgrass prairie

4.0 Abstract

Tallgrass prairie is an increasingly rare ecoregion that reaches its northernmost extent in

the province of Manitoba. There are efforts to restore and preserve this culturally and

economically significant habitat. Few insects are considered in prairie management. I examined

the effects of grazing and burning on two beneficial guilds of beetles at the Manitoba Tall Grass

Prairie Preserve (the largest remnant of tallgrass prairie in Canada). I sampled haplogastran

beetles using baited pitfall traps in sites that recently experienced burns, grazing, or no

disturbance. I found grazing supported overall greater abundance of total haplogastran beetles,

and specifically decomposer beetles. However, predatory beetles were more abundant in burned

sites. Species richness and Hill’s Shannon diversity were not affected by disturbance type.

Landscape diversity and composition were important for beetle abundance, richness, and

diversity. Rarefaction demonstrated that burned sites support more effective species than

undisturbed or grazed sites. Indicator species analysis highlighted several beetles that were

associated with one or more disturbance types. I discuss the effects of tallgrass prairie

disturbance type on often overlooked components of the beneficial prairie insect community.

4.1 Introduction

Tallgrass prairie (TGP) is the easternmost and most productive ecoregion within the

Great Plains and Canadian Prairies (Joyce and Morgan 1989). Tallgrass prairie once spanned

from Texas to southern Canada but was dramatically reduced during colonization and western

agricultural expansion (Samson and Knopf 1994). TGP was created and maintained through a

combination of large ungulate (i.e., bison) grazing, and natural and Indigenous-set fires (Pyne

1986). These processes prevented forest encroachment (Pyne 1986). The prairie’s rich, black soil

made for ideal conditions to grow crops. Largescale conversion to monoculture crops supplanted

the natural diversity of the prairie and transformed 99% of it into simplified habitats (Samson

86

and Knopf 1994, Koper et al. 2010). Southern Manitoba represents the northernmost extent of

TGP (Figure 4.1), though less than 0.1% of the former breadth of the prairie remains intact in the

province (Samson and Knopf 1994), with that number continuing to decrease (Koper et al.

2010). Rare, charismatic species occur in the Manitoba TGP, such as the eastern tiger

salamander, Ambystoma tigrinum Green, and the small white lady slipper, Cypripedium

candidum Muhl. ex Willd. Public outreach and ecotourism have led to increased funds for the

preservation of TGP through land purchases and targeted management (Costanza et al. 1997,

Veech 2006, Gerla et al. 2012, Berger et al. 2020).

Figure 4.1 Map showing the distribution of the northern tall grasslands ecoregion in North

America. Distribution is based on ecoregions as defined by the World Wildlife Fund and Olsen

et al. (2001).

TGP is often managed by mimicking historical disturbances of the Prairies, through

prescribed fires, cattle grazing, mowing, and haying (Sveinson 2003). These techniques are

chiefly applied to alter vegetation communities (Sveinson 2003). Burning the prairie reduces leaf

litter, which raises soil temperature by increasing solar absorption (Ehrenreich and Aikman

1963). Higher soil temperatures lead to an earlier growing season and greater flower production

(Ehrenreich and Aikman 1963). Faunal elements are rarely considered when planning

87

management, however periodic fires also benefit higher trophic levels in prairies (Grant et al.

2010).

Both management and landscape context can influence the positive effects that TGP can

have on the greater environment. The ecological functions provided by TGP include carbon

sequestration (Baer et al. 2002, McLauchlan et al. 2006, Adler et al. 2009, An et al. 2013), soil

erosion prevention, and water pollutant filtration (Tomer et al. 2019, Nelson et al. 2020) .

Arthropod communities also provide ecological functions in the TGP, including decomposition

and predation (Whiles and Charlton 2006). Unfortunately, arthropods are rarely considered when

planning TGP management, but see Westwood et al. (2020).

Beetle community composition is affected by environmental variables at both local and

landscape scales in North American grasslands. However, the scale of influence varies due to the

heterogenous makeup of TGP (Bossenbroek et al. 2004). Predatory beetles in restored grasslands

have greater species richness as grass cover increases at the local scale, and as semi-natural

habitats surrounding the grassland increases at the landscape scale (Crist and Peters 2014).

Predators may be more susceptible to regional (patch shape) landscape change than lower trophic

levels in the TGP (Stoner and Joern 2004). Detritivorous dung beetle species richness responds

negatively to greater landscape variety (Reynolds et al. 2018).

Haplogastra is a proposed clade that includes the proposed clades Staphyliniformia and

Scarabaeiformia (Kolbe 1908, McKenna et al. 2015) (but see Beutel and Leschen 2005).

Staphyliniformia includes the rove beetles (Staphylinidae), carrion beetles (Silphidae), round

fungus beetles (Leiodidae), feather-winged beetles (Ptiliidae), clown or hister beetles

(Histeridae), and water scavenger beetles (Hydrophilidae), among many other smaller f amiles.

The Scarabaeiformia includes scarabs (Scarabaeidae), stag beetles (Lucanidae), bess beetles

(Passalidae), and hide beetles (Trogidae), as well as several other smaller families. The

Haplogastra are thought to have evolved as detritus feeders, which is a habit and lifestyle

retained in most of its lineages (McKenna et al. 2015).

Haplogastran beetles are ideal bioindicators owing to their immense diversity (rove

beetles and scarabaeoids account for 110,000+ described species (Marshall 2018)), ease of

sampling, close associations with prey/feeding substrate, short lifespan, and low economic

importance, thus not causing concern from mass removal (Leivas and Carneiro 2012, Vieira et

88

al. 2018). However, haplogastrans are ecologically important and may have indirect economic

benefits, for example scarabs who remove cattle gastrointestinal parasite larvae from around cow

pats (Sands and Wall 2017). Haplogastrans have recently become much more common in

biodiversity studies (Hansen et al. 2018, Vieira et al. 2018). Rove beetles are increasingly

replacing carabids for environmental impact studies due to their immense diversity and close

association with certain habitats (Bohac 1999, Klimaszewski et al. 2018a).

Rove beetles are functionally diverse, but the majority are predacious or saprophagous

(Bohac 1999). Rove beetles are readily sampled using pitfall traps, especially those that are quick

running and live above the soil surface (Irmler and Lipkow 2018). Carrion beetles are likely

derived from within the Staphylinidae (McKenna et al. 2015).

The water scavenger beetles are largely aquatic, however the subfamily Sphaeridinae is

mostly terrestrial and regularly occur in cow pats and other rotting material (Short and Fikacek

2013). Hister beetles are predators that occur in dung, carrion, and rotting cacti, while others live

as inquilines in mammal nests or ant colonies, or under bark (Bousquet and Laplante 2006).

Scarabs are divided into several subfamilies, with some being largely herbivorous, while two are

dung feeders (Scarabaeinae and Aphodiinae). Scarabaeinae tend to be “rollers”, or “tunnelers”,

either rolling pieces of dung away to a separate location, or tunneling under the dung to

provision their larvae (Helgesen and Post 1967, Halffter and Edmonds 1982). The Aphodiinae

are “dwellers” that complete their entire life cycle within dung or rich humus source (Helgesen

and Post 1967, Halffter and Edmonds 1982).

My goal was to determine the effects of prairie disturbance type and landscape factors on

predacious and saprophagous (decomposer) beetles. I used baited pitfall traps to capture beetles

in TGP sites that differed in their history of disturbance. My three main objectives were to: i)

determine the effects of fire and grazing disturbance on predator and decomposer haplogastran

beetle communities, ii) examine how landscape scale and local scale features relate to diversity

of these haplogastran beetles and iii), determine how disturbance, landscape level, and local scale

factors differentially affect decomposers and predators. I predict that grazed sites will have

higher haplogastran diversity and abundance compared to burned sites due to increased quality

and abundance of detritus (e.g., remnants of cow dung). Grazed grasslands have faster

decomposition rates than burned ones (Hofstede 1995), which I speculate is related to abundance

89

and diversity of decomposer beetles. I predict that TGP with lower landscape diversity will have

a greater diversity of decomposer beetle captures. I predict that increasing grass cover will

benefit abundance and diversity of predacious beetles (Crist and Peters 2014). With an increasing

shift towards understanding whole communities and ecosystem processes, the underlying aim in

conducting this study is to catalogue an often overlooked but immensely diverse component of

the prairie fauna, as well as attempt to uncover patterns that may protect them and the valuable

prairie services they provide.

4.2 Methods

4.2.1 Site selection

I chose 18 sites in the Manitoba Tall Grass Prairie Preserve, the largest remnant tallgrass

prairie in Canada (headquartered at 49.1548, -96.7299). Sites were chosen based on landcover

data and management history provided by Nature Conservancy Canada (NCC) and its partners.

The preserve is surrounded by agricultural land and is divided into a North and South block.

Sites were chosen in the North block, as management strategies were more discreet (grazed or

burned, not both) and it allowed for multiple sites to be easily visited quickly. The North block is

a relatively wet, boggy prairie, with the water table being near the surface until about the end of

June. My sites fell into three distinct categories: i) grazed sites (n=6) have had cattle on them at

least once since 2007, but were not burned in at least 24 years, the ranges of recent grazes were

one site that was grazed in 2006 and 2007, one site that was grazed annually from 2008-2011,

two sites were grazed in both 2011 and 2012, and two sites were grazed annually from 2012-

2016. One site was also grazed during my 2018 field season, however my study area had an

enclosure around it to prevent disturbance; ii) burned sites (n=5) experienced a burn at least once

since 2012 but have not been grazed in at least 24 years, with ranges of four sites being burned in

an unplanned wildfire in 2012, and one site experiencing a controlled burn in 2016; and iii)

undisturbed sites (n=7) that have not been grazed or burned in at least 24 years.

90

Figure 4.2 Map of the distribution of the study sites relative to the town of Stuartburn. Burned

sites are represented by red triangles, grazed sites by green circles, and undisturbed sites by blue

squares. The insert shows an example of the Latin square design of the baited pitfall traps, with

gingerbread men representing ginger baited traps, cows representing cow dung traps, apples

representing EtOH + amyl acetate traps, and fish representing fish skin baited traps.

At each site, I deployed 16 pitfall traps in a 4 x 4 matrix with a minimum separation of 15

m. Pitfall traps are frequently used in biodiversity studies, and are effective at catching ground

dwelling beetles (The Biological Survey of Canada 1994). Four distinct baits were chosen to

attract haplogastran beetles to the pitfall traps: i) 30 mL of cow dung, ii) 30 mL of 99:1 mixture

of 95% ethanol: amyl acetate (Anderson and Nelson 1993), iii) 30 g of cut and peeled ginger

root, and iv) 30 g of walleye (Sander vitreus (Mitchill)) skin and scales. Four of each bait were

used at each site arranged in a Latin square design. This design ensures that each bait appeared

only once in each row and column of my sampling grid (Figure 4.2). Dung was obtained fresh

from dairy cows at Glenlea Research Station, south of Winnipeg. It was put into vials and frozen

until use. Amyl acetate attracts detritivores including some flies and haplogastran beetles

(Anderson and Nelson 1993, White 1998). Ginger was obtained in bulk from a local

supermarket. Ginger mimics the smell of wood volatiles and has been shown to successfully

attract stag beetles (Harvey et al. 2011). Walleye skin and scales were provided by Freshwater

Fish, a federal government agency located in Winnipeg, Manitoba.

The baits were placed in 50 mL centrifuge tubes and placed within a 1 L plastic cup filled

with 2–3 cm of 1:1 propylene glycol: water mixture. This cup was then placed in a second cup in

the ground such that both cups together were flush with the ground. A metal grid with 2.54 2 cm

N 49.18

N 49.10

W 96.74 W 96.65

91

holes was then placed overtop to try and exclude vertebrates from accidentally falling into the

pitfall. A wooden rain cover was placed 2–3 cm above the trap to prevent the trap from

overflowing.

Samples were taken in a way that one of each group (grazed, burned, and undisturbed)

were sampled at the same time, except for one group that consisted of a grazed and two

undisturbed sites. In both years, each site was sampled three times. In the summer of 2017, the

traps were run for two days, with the baits taken out of the freezer in the morning, and placed in

the field before noon, and then the captures collected 48 hours later. In 2018, traps were run for

one week to increase the number of beetles captured. Baits were removed from the freezer the

night before deployment to ensure they were attractive as soon as they were put in the field. The

traps were run for longer in the second year to attract a more diverse and abundant beetle

assemblage, as the majority of rove beetles arrive at dung after the third day (Guimarães and

Mendes 1998), and at decomposing flesh after the fourth day (Cruise et al. 2018). Once the traps

were retrieved, beetles were frozen until they were washed with soapy water and then stored in

70% ethanol. Beetles were then dried and pinned or point mounted. Specimens which could not

be placed to species were sent to the Canadian Museum of Nature (2017 scarabs), the Canadian

National Collection of Insects, Arachnids, and Nematodes in Ottawa, Ontario (2017 captures,

2018 rove beetles) for expert identification, or to Carleton University in Ottawa, Ontario (2018

round fungus beetles). Specimens are deposited in the J.B. Wallis-R.E. Roughley Museum of

Entomology at the University of Manitoba. Duplicate vouchers are deposited at the Canadian

National Collection and at the Canadian Museum of Nature in Ottawa.

4.2.2 Landscape Mapping

Landcover data surrounding my sites was obtained via a usage agreement with Nature

Conservancy Canada and their partners. Areas adjacent to the preserve were hand digitized based

on satellite imagery. I extracted radii of 250, 500, 1000, 1500, and 2000 m p laced around the

centre of each site in ArcMap v.6.2 (ESRI 2018). Polygon data were extracted to determine what

land use types dominated the varying ranges of buffers. Land cover was grouped into the

following categories: forest, peri-urban, structure, savannah, water, swamp, prairie, pasture, crop,

river, edge (ditch), other, and road. These categories were created based on what is readily

visible in satellite imagery and NCC categories, though the latter were largely combined into

92

single categories, such as different prairie subcategories and different forest types. Patch Analyst

v. 5.2 (Rempel et al. 1999) was used to calculate Shannon diversity of the landscape and the

mean shape index for each buffer radius. Shannon diversity increases as different patch types are

added to the landscape and is maximized when all patch types are of equal proportion (Rempel et

al. 1999, McGaril & Marks 1994). The mean shape index starts at one, a situation in which all

patches are perfectly circular, and increases as the complexity of patch shapes increase (Rempel

et al. 1999). Individual landcover proportions were also extracted. These portions were then put

in to a principal coordinate analysis (PCA) using the ‘FactoMineR’ package (Husson et al.

2020). The first principal coordinate axis was used for subsequent models. In 2018, 1 x 1 m

quadrats were also used to measure percent cover of bare ground, grass, and forb/woody cover

and number of inflorescences (intended for a parallel study examining bees). Quadrats were

tossed overhead 16 times twice in 2018. The average of the 16 times was recorded, and the

average of both rounds was used per site. A PCA was then performed to extract the first axis

coordinate for each site (Figure 4.10).

4.2.3 Identification/Data analysis

Haplogastran beetles were identified using reference collections and several published

keys (Helgesen and Post 1967, Campbell 1968, Smetana 1971, Campbell 1979, Watrous 1980,

Campbell 1983, 1984, Anderson and Peck 1985, Smetana 1988, Campbell 1991, Baranowski

1993, Campbell 1993a, 1993b, Smetana 1995, Peck and Stephan 1996, Newton et al. 2001, Peck

and Cook 2002, Bousquet and Laplante 2006, Gordon and Skelley 2007, Peck and Cook 2009,

Brunke et al. 2011, Peck and Cook 2013). Beetles were analysed both together and separated by

feeding guild (predator and decomposer). Feeding guild was established via published studies

(Helgesen and Post 1967, Anderson and Peck 1985, Smetana 1988, Dennis et al. 1990, Good and

Giller 1991, Jo and Smitley 2003, Bousquet and Laplante 2006, Majka and Klimaszewski 2008,

Thayer 2016). Omnivores were included in both guilds, and species with unknown feeding

habits/non decomposer or predator beetles were excluded from both. All statistical analyses were

performed in R (R Core Team 2017). Beetle community composition was examined using

distance based redundancy analysis (Legendre and Anderson 1999) using the ‘capscale’ function

of the ‘vegan’ package (Oksanen et al. 2018). For the redundancy analysis, the 2017 and 2018

93

data were analysed separately. I used permutational multivariate analysis of variance

(perMANOVA) to test for effects of disturbance type on beetle communities using the ‘adonis’

function in the ‘vegan’ package (Oksanen et al. 2018). I used the Bray-Curtis dissimilarity for

community composition analyses.

Mixed-effects models with site as a random effect were used to test for differences in

abundance and diversity between disturbance types. I used the first two Hill numbers—species

richness and the Hill’s Shannon index—as measures of diversity (Jost 2007). Hill numbers

include species richness (q=0), Hill’s Shannon effective species number (q=1), and the Inverse

Simpson’s effective species number (q=2). These differ in the amount of weight placed on

species abundance, with species richness being independent of abundance, the Hill’s Shannon

being weighted toward rare species, and the Inverse Simpson’s being weighted to more abundant

species (Chao et al. 2014). The ‘q’ numbers are derived from Hill (1973), who used the notation

devised by Rényi (1961).

I used a model selection approach using Akaike Information Criterion with a small

sample size correction (AICc). Models were built gradually, first by including average

temperature, average precipitation, day of year, and year. Then, the model was dredged using the

‘dredge’ command of the ‘MuMIn’ package to determine which variables to keep going forward

(Sakamoto et al. 1986, Barton 2018). Based on early model selection, I determined that the

beetles responded most to the 500 m buffer size, which is also supported by other investigations

on predatory and decomposer beetles in grassland ecosystems (Liu et al. 2014), so this scale was

used in subsequent models. Landscape metrics, disturbance, and groundcover variables were

added to the base model. I then compared my global model to simpler models and selected the

best model using the AICc. Preferred models were assessed using the ‘check_model’ f unction of

‘performance’ package (Ludecke et al. 2020) to determine how well the models fit assumptions

and their distributions, and whether collinearity between fixed effects was present. Abundance

models were initially built with a negative binomial distribution, species richness models with a

quasi-Poisson, and Hill’s Shannon models with a Gamma distribution. However, models were

re-built under an alternate distribution (such as the Gaussian) if the package ranked it as better

performing using the ‘compare_performance’ command. In cases where the Gaussian

94

distribution was used, a Shapiro-Wilk test was performed on the residuals to test for the

assumption of normality (Royston 1982a, 1982b, 1995).

Indicator species analysis was performed to test if certain species were strongly tied to a

certain disturbance type and site(s) (De Cáceres and Legendre 2009, De Cáceres et al. 2010, De

Cáceres and Jansen 2016). Two methods were used, both from the ‘indicspecies’ package (De

Cáceres and Jansen 2016); the ‘Indval.g’ method, which stands for the group equalized indicator

value and does not consider absences from sites as being evidence against a species being an

indicator, and the ‘r.g.’ method, which stands for the group equalized point biserial correlation

and considers abundances outside of the group which a species may be an indicator for as weight

towards whether or not to classify a species as an indicator (De Cáceres et al. 2010). The ‘.g’

suffix allows for assumed variability within disturbance types to be proportional to the group size

(De Cáceres et al. 2010). Indicator species analysis was performed by ‘Disturbance Type’, the

focus of my study, as well as by ‘Site’, which used the six samples as replicates. I felt this would

be useful information for land managers in case rare/endangered/new species are among the

indicators, which could lead to more careful or more targeted management of certain sites.

Species diversity is sensitive to sample size. Rarefaction allows for fair comparison

between samples with different abundances and different sampling efforts. I generated rarefied

diversity values using Hill numbers for each site type by collapsing each individual sample into

its disturbance type. I used the package ‘iNEXT’ (Hsieh et al. 2019) to generate graphs showing

the interpolated and extrapolated rarefied diversity (represented by q=0 (species richness), q=1

(Hill’s Shannon effective species number), and q=2 (inverse Simpson’s effective species

number) of the three disturbance types.

4.3 Results

I caught a total of 19,448 haplogastran beetles over my two years of study, including

10,483 in the rove beetle subfamily Aleocharinae, which were not identified further. The

remaining 8,965 specimens represented 124 species from 70 genera. There were 1,351 predatory

beetles caught representing 69 species. There were 7,628 decomposer beetles from 55 species.

95

Of these two guilds, eight were omnivorous species, accounting for 64 specimens. Besides the

aleocharines, there were eight species totaling 50 specimens that could not be placed into my

predator or detritivore categories. There were 3,634 (1,931 non aleocharine) beetles caught on

burned sites, 9,293 (3,659 non aleocharine) caught on grazed sites, and 6,521 (3,375 non

aleocharine) caught on undisturbed sites. In 2017, when traps were left out for only 48 hours,

853 (483 non aleocharine) beetles were caught. In 2018, when traps were left out for 7 days,

18,595 (8,482 non aleocharine) beetles were caught. Cow dung baits caught 1,825 beetles (474

non aleocharine), EtOH + Amyl Acetate baits caught 2,025 (928 non aleocharine) beetles, ginger

root caught 2,372 (440 non aleocharine) beetles, and fish skin caught 13,226 (7,123 non

aleocharine) beetles. Nineteen species represent new provincial records for Manitoba, including

one likely undescribed species of Stenus.

4.3.1 Community Composition - Distance-based Redundancy Analysis/perMANOVA

The total haplogastran beetle communities (including the aleocharine captures) differ

between the three categories of disturbance, but communities of predatory and decomposing

beetles did not differ. Disturbance constraints on the dbRDA explained 25% of the variation in

2017, and 24% of the variation in 2018 (Table 4.1, Table 4.2).

Table 4.1 Summary of amount of variation explained by constrained (Disturbance type) factors

in dbRDA for the total beetle communities in both years.

Inertia Proportion

2017 - Total 3.5602 1.000

2017 – Constrained 0.8993 0.2526

2017 - Unconstrained 2.6609 0.7474

2018 - Total 1.876 1.000

2018 - Constrained 0.454 0.242

2018 - Unconstrained 1.422 0.758

96

Table 4.2 Summary of the effect of ‘Disturbance Type’ on beetle dissimilarities in distance -

based redundancy analyses using the ‘anova’ function with ‘by= “terms”’ specified. Results were

obtained by running 999 permutations.

Community Year F model p value

All haplogastran

beetles 2017 2.5348 0.006

Predator beetles 2017 0.8057 0.759

Decomposer beetles 2017 1.4695 0.111

All haplogastran

beetles 2018 2.3928 0.012

Predator beetles 2018 1.3342 0.091

Decomposer beetles 2018 1.2154 0.264

97

Figure 4.3 Distance-based redundancy ordination of all 2017 haplogastran beetles (left) (F = 2.5348, p = 0.006), predatory beetles

(centre) (F = 0.8057, p = 0.759), and decomposer beetles (right) (F = 1.4695, p = 0.111), constrained by disturbance type using the

Bray-Curtis method. The three sampling rounds were combined in to one value per site to avoid pseudo-replication. The red triangles

represent ‘burned’ sites, green circles represent ‘grazed’ sites, and blue circles represent ‘undisturbed’ sites. Ellipses re present 95%

confidence intervals around the disturbance types.

98

PerMANOVA allowed for numerical post-hoc comparison among disturbance groups

(Table 4.3). Grazed site beetle communities were significantly different from burned sites in both

years (Table 4.3, Figure 4.3, Figure 4.4), but only differed from undisturbed sites in 2017, though

this was confounded by the fact that they had unequal dispersion, though as Figure 4.3 shows,

the differences are likely real. In 2017, grazed and undisturbed sites had significantly different

dispersions, but they also show little overlap in the ordination. The predatory and decomposer

communities did not differ among disturbance categories.

Table 4.3 Summary of perMANOVAs performed on the Bray-Curtis dissimilarity of different

beetle communities in 2017 & 2018. The Post-hoc column represents the results of a Tukey’s

post-hoc test with a Bonferroni correction. B = ‘Burned’, G = ‘Grazed’, U = ‘Undisturbed’. *

refers to an instance where disturbance types were determined to have unequal dispersion

parameters, thus this result should be interpreted with caution.

Community Year F model R2 p value Post-hoc

All haplogastran beetles 2017 2.5375 0.2528 0.007

B-U: 1

B-G: 0.018

G-U: 0.042*

Predator beetles 2017 0.6983 0.08518 0.836 N/A

Decomposer beetles 2017 1.4906 0.16579 0.096

B-U: 1

B-G: 0.264

G-U: 0.288

All haplogastran beetles 2018 2.4721 0.2479 0.018

B-U: 1

B-G: 0.012

G-U: 0.066

Predator beetles 2018 1.3514 0.15267 0.1

B-U: 0.564

B-G: 0.039

G-U: 1

Decomposer beetles 2018 1.2082 0.13874 0.278 N/A

99

Figure 4.4 Distance-based redundancy ordination of all 2018 haplogastran beetles (left) (F = 2.3938, p = 0.012), predatory beetles

(centre) (F = 1.3342, p = 0.091), and decomposer beetles (right) (F = 1.2154, p = 0.264), constrained by disturbance type using the ‘Bray-Curtis’ method. The three sampling rounds were combined in to one value per site to avoid pseudo -replication. The red triangles represent ‘burned’ sites, green circles represent ‘grazed’ sites, and blue circles represent ‘undisturbed’ sites. E llipses represent 95% confidence intervals around the disturbance types.

100

4.3.2 Models

4.3.2.1 Disturbance effects

When all haplogastran beetles are treated together, disturbance type has a significant

effect on abundance (Figure 4.5). Grazed sites host a significantly greater abundance of beetles

than either burned or undisturbed sites. Disturbance type was not a significant variable in the best

model for species richness, however it was for the Hill’s Shannon model, where I see the

opposite effect of disturbance than abundance, where burned sites host significantly more

effective species than grazed or undisturbed sites (Figure 4.5).

When beetles were separated based on feeding guild, only abundance was affected by

disturbance type. Burned sites had greater abundance of predatory beetles than grazed and

undisturbed sites (Figure 4.6). Conversely, grazed sites support the greatest abundance of

decomposer beetles, having significantly more than undisturbed sites, but not significantly more

than burned sites (Figure 4.6). When looking at aleocharine abundance alone, grazed sites had

significantly more aleocharines than burned or undisturbed sites (Figure 4.7).

101

Figure 4.5 Effect plot of the predicted effect of disturbance on haplogastran beetle abundance (left), and on the effect of Hill’s

Shannon effective species number (right) when all other variables part of the best fitting GLMMs are held at their average or default.

Different letters above the disturbance types represent significantly differences based on GLMMs (p<0.05). Points represent the

median value of the predictor variable of the disturbance type subset, and whiskers represent 95% confidence intervals.

102

Figure 4.6 Effect plot of the predicted effect of disturbance type on predatory haplogastran beetle (left), and decomposer beetle (right)

abundance when all other variables part of the best fitting GLMMs are held at their average or default. Different letters abo ve the

disturbance types represent significantly differences based on GLMMs (p<0.05). Points represent the median value of the predictor

variable of the disturbance type subset, and whiskers represent 95% confidence intervals.

103

Figure 4.7 Effect plot of the predicted effect of disturbance type on aleocharine beetle

abundance when all other variables part of the best fitting GLMM is kept at their average or

default value. Different letters above the disturbance types represent significantly differences

based on a GLMM (p<0.05). Points represent the median value of the predictor variable of the

disturbance type subset, and whiskers represent 95% confidence intervals.

4.3.2.2 Landscape effects

The PCA for the landcover types at 500 m was part of the best model for combined beetle

abundance, where it had a positive effect. The first axis of the PCA was related to forest, ditch,

and pasture cover on beetle abundance, with a negative effect of increasing prairie and swamp

cover (Figure 4.9). This positive effect was also observed in the species richness, and Hill’s

Shannon models.

104

No landscape factors were recovered in the best models for predatory beetles or

aleocharine beetles. The preferred model for decomposer abundance included positive effects of

the landcover PCA and the landscape mean shape index. The latter implies that increased patch

shape complexity and greater amount of edge in the landscape has a positive effect on

decomposer beetle abundance. Landcover PCA was also included in the best model for

decomposer beetle species richness, again demonstrating a positive effect along the first axis. No

landscape factors were part of the best model for decomposer beetle Hill’s Shannon effective

species number.

4.3.2.3 Ground cover effects

The ground cover PCA (Figure 4.10) was part of the best model for combined beetle

species richness, where it had a negative effect, implying that lesser amounts of forbs, and

greater amounts of grass cover were beneficial to beetle species richness.

The ground cover PCA was also part of all three predatory beetle models, again showing

a negative effect of forbs, and a positive effect of increased grass cover on abundance, richness,

and diversity. Ground cover was not part of any of the decomposer beetle models or the

aleocharine abundance model.

4.3.3 Rarefaction Curves

Burned sites had greater rarefied effective species than grazed and undisturbed sites using

Hill numbers q=1 and q=2. This pattern was present in the combined beetle (with aleocharines

excluded) and the decomposer beetle data, but not present for predatory beetles (Figure 4.8).

Undisturbed sites also demonstrated significantly greater effective species numbers compared to

grazed sites in the same pattern. Undisturbed sites support greater species richness than grazed

sites for predatory beetles.

105

Q=0

Q=0

Q=1

Q=1

Q=2

Q=2

S

pec

ies

Div

ersi

ty

S

pec

ies

Div

ersi

ty

106

Figure 4.8 Rarefied diversity curves of haplogastran beetle captures (except the subfamily Aleocharinae), with the top row

representing all beetles combined, the middle row representing predatory beetles, and the bottom row representing decomposer

beetles. The columns represent the q numbers, where left represents q = 0 = species richness, centre represents q = 1 = Hill’s Shannon

effective species number, and right represents q = 2 = Inverse Simpson’s effective species number.

S

pec

ies

Div

ersi

ty

Q=0 Q=1 Q=2

107

4.3.4 Indicator Species Analysis

The two methods used for the indicator species analysis returned some of the same

beetles as indicators for certain disturbance types but differed for others. Burned sites have four

indicator species, two of which are decomposers, and two of which are predatory. Bledius

strenuus Casey feeds on algae (Thayer 2016), which were potentially more abundant on the

surface in burned sites due to greater light availability and less trampling on their burrows.

Grazed sites have four indicator species, two decomposers and two predators. Two of the

indicator species of burned and grazed sites are shared between the two, implying that they may

be indicators of disturbance in general. Undisturbed sites did not have any indicator species

(Table 4.4).

Table 4.4 Results of Indicator Species Analysis on beetle species by disturbance type. Guild

refers to feeding guild that beetle belongs to, with P = predator and D = decomposer. Analyses

were run for 999 permutations. Only indicators with more than 10 specimens caught are here

reported.

Species Guild # Disturbance Type Method Statistic p-value

Bledius strenuus

Casey D 19 Burned

Indval.g,

r.g.

Indval.g =

0.441

r.g. = 0.334

Indval.g =

0.001

r.g. = 0.004

Anotylus niger

(LeConte) P 72 Grazed Indval.g 0.477 0.047

Heterosilpha

ramosa

(Say)

D 261 Burned +

Grazed Indval.g 0.588 0.036

Neobisnius

jucundus

(Horn)

P 74 Burned +

Grazed

Indval.g,

r.g.

Indval.g =

0.517

r.g. = 0.301

Indval.g =

0.002

r.g. = 0.003

Philonthus

schwarzi

Horn

P 128 Burned r.g. 0.262 0.022

108

Nicrophorus

hebes Kirby D 1,044 Grazed r.g. 0.297 0.008

Indicator species was also run by site instead of disturbance type and the results are

included in the supplementary material (Tables 4.9, 4.10).

4.4 Discussion

The intensity and frequency of fire may play a role in determining the effect on arthropod

communities. A study in Manitoba TGP found that optimal fire season for promoting diversity

and evenness differed between spiders and the floral community (Wade 2002). Burn season had

no effect on ground beetle communities (Roughley et al. 2010). While the literature is scant for

beetle fauna, studies on open habitat butterflies provide evidence that single, infrequent, high

intensity wildfires are more beneficial than maintenance prescribed fires for butterfly

conservation (reviewed in Swengel 2001).

Though studies suggest that burn management history has negligible effects on

decomposition rates in TGP (Reed et al. 2009), drought conditions, like those experienced in

2018, do start to produce decomposition differences between burn/cessation regimens (Reed et

al. 2009). Whether abundance or diversity of decomposer haplogastran beetles influence this

process is unclear, however studies in agricultural settings suggest that low organic nutrient

availability at the surface in summer conditions lead to decreased numbers of rove beetles

(Brunke et al. 2014).

Other studies have noted detrimental effects on a coprophagous haplogastran beetle and

dung removal rate via presence of predatory haplogastran beetles (Wu et al. 2011), which could

potentially explain my large differences in decomposer and predator beetle abundances in my

different disturbance types. This same experiment saw negative effects of predators on plant

biomass, implying that sites with large amounts of predators could negatively impact the floral

component of grasslands (Wu et al. 2011).

The results of the dbRDA and perMANOVA imply that the type of disturbance can

greatly influence the total beetle community composition. This has important implications for

109

managing the prairie going forward. Ideally, it would be useful to compare these communities to

sites that are managed with a combination of burning and grazing to see if the combined

approach is intermediate between the two disturbance types, or if it creates a distinct assemblage

of its own.

Low intensity grazing, which increases soil compaction, has previously been reported to

increase rove beetle abundance (Hofmann and Mason 2006). However, rove beetle diversity

decreased with increasing soil moisture levels (Hofmann and Mason 2006), and this experiment

had confounding factors that the authors warn reduces the ability to generalize their results. This

exemplifies the need to measure soil variables if a similar experiment were to be conducted again

given that compaction, nutrient availability, soil acidity, and organic matter availability are all

known to be affected by features and age of burning and grazing in TGP, which would likely

also affect all higher trophic levels (reviewed in Faber 2008, A. R. Westwood (pers. comm.)

2021). Large herbivores can have detrimental effects on dung feeding insects through the use of

antihelminth medications (Wall and Strong 1987, Madsen et al. 1990, van Klink et al. 2015),

which could presumably also affect the predatory insects that feed on them. Unfortunately,

medication regimens of nearby farms that utilize the preserve were not considered in the

planning of this project. As well, trampling, even minimally, on grasslands has previously been

shown to negatively impact arthropod communities to a much greater degree than grassland plant

communities (Duffey 1975, van Klink et al. 2015). Grazed sites showed highest decomposer

abundance, but lowest rarefied decomposer diversity. Grazed sites did not however have lower

rarefied diversity of predatory beetles, which could suggest that increased decomposer

abundance can lead to increased predator diversity on grazed sites, which has been hypothesized

before (van Klink et al. 2015).

There is evidence that smaller bodied beetles, which represent the bulk of the diversity in

my samples, but not so much the bulk of abundance in my study, fare poorer in grazed

grasslands compared to a lesser effect on larger bodied beetles (Dennis et al. 1998). Similar to

my findings in showing negative impacts of grazing, cattle grazing on grasslands in Europe has

been shown to negatively impact insect diversity compared to long undisturbed sites (Kruess and

Tscharntke 2002). However, they also found reduced insect abundance at intensively grazed

sites, which conflicts with my findings, though they were not sampling in the same manor and

110

for the same functional groups that I was (Kruess and Tscharntke 2002). A study on coastal salt

marshes in the UK found significantly higher rove beetle abundance in several months on a site

with low intensity cattle grazing and an elevated water level (Hofmann and Mason 2006). They

also found evidence that suggests that rove beetle species turnover happens quickly following a

site shift from grazing to non-grazing (Hofmann and Mason 2006). Since most of my sites had

been a year or more since active grazing, this implies that my results may be interpreting the

outcome of a cessation of grazing rather than the direct effect of grazing, especially since no

dung feeding scarabs were recovered as indicators of grazed sites. Like my study, Doxon et al.

(2011), in a prairie system that historically was shaped by burning and grazing, found that

beetles were more abundant in traditional, graze only sites than those exposed additionally to

fire. Carrion-associated rove beetle species composition is known to be affected by

intensification of management, with generalist species replacing rare/specialist species under

increasing management intensity (Weithmann et al. 2020). This could explain my findings of

community composition differences between grazed and burned sites if cattle grazing has a more

intense effect on the rove beetle fauna in the TGP.

Herbaceous dicot forbs are the main source of plant species richness in TGP (Whiles and

Charlton 2006). Woody encroachment has known negative effects on the diversity of these forbs

(Taft and Kron 2014). Beetle family richness has been shown to be positively correlated to native

prairie plant abundance (Jonas et al. 2002). Increased plant diversity in grasslands increases

arthropod diversity (Siemann et al. 1998). As well, herbivorous arthropod diversity was also

codependent on predacious invertebrate diversity (Siemann et al. 1998), implying that

management that promotes greater plant diversity can have cascading effects on the diversity of

several invertebrate feeding guilds. Though increased litter layer, which would be associated

with my undisturbed sites (not measured, pers. obs.), has been shown to increase predatory

arthropod abundance (Langellotto and Denno 2004, van Klink et al. 2015), I did not see this

pattern in my results.

Studies on ground beetles, a predominantly predatory beetle guild, in TGP have shown

negligible effects of frequent burning on community assemblage (Cook and Holt 2006). The

authors note however, that as ground beetles are comparatively unspecialized predators, their

conclusions cannot be extrapolated to other predatory guilds (Cook and Holt 2006). A study in a

111

grassland ecosystem found that livestock stocking rate negatively affected aggregation of a

predatory rove beetle, thought to possibly be due to less plant litter on grazed sites (Dennis et al.

2002). Some studies have found increased predatory beetle abundance on open sites during the

recovery phase post-fire (reviewed in Swengel 2001). As well, reduced litter levels following a

fire are known to reduce decomposer abundance (Seastedt 1984, Seastedt et al. 1986, Kral et al.

2017), while simultaneously increasing predatory beetle abundance (Tester and Marshall 1961).

This is consistent with my findings of increased predator fauna on my burned sites since all but

one experienced fire several years prior, as well as my increased decomposer numbers on my

grazed sites.

The trampling associated with grazing can have direct lethal effects on large dung beetles

(Negro et al. 2011, van Klink et al. 2015), and though herbivores supply the main food source to

many decomposers in the form of dung, studies suggest that high stocking rates can have

negative effects on dung beetle diversity and abundance in temperate grasslands (Negro et al.

2011, van Klink et al. 2015). Some have suggested that larger insects respond more positively to

the grass cover heterogeneity that grazing produces than smaller insects (Klink et al. 2013). This

could potentially help explain my findings of greater abundance of decomposers, many of which

were large species such as Nicrophorus spp. and Necrophila spp., on my grazed sites compared

to the other disturbance types. It could also explain the decreased richness on my grazed sites if

the smaller species, which made up the bulk of the species richness, respond more negatively to

grazing. It is has been suggested that decomposer beetles may be less abundant on burned sites

due to a lack of vegetation cover, leading to quick desiccation times of the carrion and dung that

they would typically feed on (Beucke 2009). Whether this theory is driving my findings is not

clear.

Aleocharines were assumed by me to be much more abundant on my grazed sites due to

large numbers of Aleochara spp., which are parasitoids on dung and carrion feeding Diptera

(Klimaszewski 1984), though I did not identify the genus of any of my aleocharine captures, and

it could instead be that grazed sites supported more abundance of aleocharines that depend on

organic matter that may be in greater supply on sites free from burns. This could include a

hidden source of diversity that would augment the species richness of grazed sites, as

demonstrated by recent studies in Manitoba recovering 120 species of aleocharine beetles present

112

in the province (Klimaszewski et al. 2018b), however the number of grassland associated species

is much less than this, and without ID confirmations, I do not wish to speculate.

The greater amount of indicator species for the burned and grazed disturbance compared

to no indicators for undisturbed sites may hint that these beetle communities evolved in tandem

with the TGP to respond to frequent disturbance, however the species recovered have broader

distributions in Canada (Bousquet et al. 2013), so perhaps they are adapted to disturbance in

broader ecoregions. At the site level, several undisturbed sites were important for rare species,

implying that refugia should be maintained when possible.

There are conflicting results as to how decomposer beetles respond to landscape

diversity, with some studies finding a negative effect of increased diversity on richness

(Reynolds et al. 2018), while others find beneficial effects of increased landscape diversity on

beetle diversity (Rös et al. 2012). My study did not corroborate the findings of Liu et al. (2014)

who found a positive effect of landscape diversity on decomposer beetle abundance. My results

suggest that landscape diversity in TGP does not affect decomposer beetle abundance.

A previous study in Manitoba TGP found that predatory arthropods were more abundant

in the forest ecosystem adjacent to TGP (Roughley et al. 2006), which could explain the positive

effect of forest cover on my captures, though the pattern was not seen in my exclusively

predatory model. Increased pasture area has been implicated as increasing rare dung beetle

species richness in grazed ecosystems (Buse et al. 2015), which could be contributing to an

increase in species in areas with more pastural area in the environment. Though dung beetles

(Scarabaeidae: Scarabaeinae, Aphodiinae) made up only a minor part of my decomposer fauna,

they are known to exhibit intermediate characteristics of forest and pasture communities when

both landscapes are present in the surrounding environment (Somay et al. 2021). The presence of

both these habitats surrounding the prairie could lead to increased diversity, which could in turn

promote the many positive functions performed by the dung beetle community. However, a

study in Europe, where many dung beetles are native and tied to longstanding animal agricultural

practice, found that dung beetle richness severely declined in reclaimed forest sites compared to

preserved agricultural grasslands (Errouissi and Jay-Robert 2019). Another study examining the

differential response of dung beetle communities to different habitat types showed that dung

beetle abundance was maximized in mature, late succession forest, but richness and diversity

113

were maximized in pastureland (Tocco et al. 2013). This could explain why the landcover

variable has a positive effect in both my abundance and diversity models, where both pasture and

forest sit in the positive aspect of the first axis of the PCA, however it must be stressed that dung

beetles made up a fairly small portion of my decomposer captures. Studies on rove beetles have

revealed positive effects of forest and forest edge in the landscape on abundance and richness

compared to grassland environments (Tóthmérész et al. 2014). This could explain my positive

effect of the amount of forest in the environment, and the negative effect of the amount of prairie

in the environment according to my landcover PCA, however the increased diversity may not

strictly be the goal of conserving the TGP if aiming to specifically conserve grassland associated

species. Contrary to this, however, is the finding that increased encroaching forest cover in

prairie grasslands lead to significant decreases in silphid abundance (Walker Jr and Hoback

2007).

My findings of decomposers responding positively to patch shape complexity contrasts

with another study that found that dung beetle richness decreased with increasing shape

complexity (Rocha-Ortega and Coronel-Arellano 2019), though my study areas are quite distinct,

and the taxonomic scope of my study is much broader, with dung beetles forming only a small

component. My results contrast to a previous study which found a negative effect of increasing

patch shape (more edge) on generic diversity of ground-dwelling arthropods, as well as

decreased richness of beetles compared to low edge patches (Orrock et al. 2011). Some studies

have found that decomposer beetles respond to landscape composition (Verdú et al. 2011,

Sánchez-de-Jesús et al. 2016, Rocha-Ortega and Coronel-Arellano 2019), sometimes pointing

out that they respond at landscape scales more than local cover scales (Rocha-Ortega and

Coronel-Arellano 2019), which is consistent with my findings, but other times pointing out that

both landcover and local patch attributes affect composition (Sánchez-de-Jesús et al. 2016).

Predator beetles, however, are known to respond to local ground cover, including positive effects

of increased grass cover on predatory beetle richness in grassland ecosystems (Crist and Peters

2014), which my results support.

Soil and vegetation characteristics have been implicated as being the most informative

variables explaining rove beetle community structure (Hofmann and Mason 2006), and diversity

(Weithmann et al. 2020). Soil characteristics are also known to influence the amount of

114

ecological functions performed by dung beetle communities (Griffiths et al. 2015, Nunes et al.

2018), and the abundance of silphid communities (Jakubec and Růžička 2015, von Hoermann et

al. 2018). This makes it clear that these factors should be measured in a follow up study.

Increasing plant species richness is known to increase insect richness in TGP (Zavaleta et al.

2010), thus it would also be useful to measure overall plant diversity in follow up studies, even if

it seems that it would likely have little effect on predator and decomposer communities.

Unfortunately, the sheer abundance and morphological homogeneity of the aleocharines

(Staphylinidae: Aleocharinae) prevented us from identifying them beyond subfamily. They have

however been stored in case specialists wish to identify them for a re-examination of the effects

of disturbance type and landscape on TGP beetles down the road. There are notable limitations to

understanding my results in a large context, as most previous work on decomposers has focused

on dung beetles, often in very disparate landscapes. While scarabs were part of my captures, they

were far from the most abundant group. There have also been some studies on silphids, however

not many in the prairies, with the exception of investigations into the rare and endangered

American burying beetle, Nicrophorus americanus (Olivier).

Some studies suggest that it is the relative proportions of few efficient players that drive

decomposition as opposed to overall diversity (Tonelli et al. 2020). However other studies have

shown a clear effect of species richness, as opposed to increased abundance, on measured

decomposition (Farwig et al. 2014, Nunes et al. 2018). Some studies have suggested a middle

ground of measuring beetle functional diversity as opposed to relying on taxonomic diversity

(Griffiths et al. 2015). A general consensus, based on a broad meta-analysis, points to negative

implications of biodiversity and individual species loss, brought on by climate change, on the

process of decomposition (Hooper et al. 2012). Therefore, it is imperative that studies in the TGP

incorporate an ecosystem function aspect to future experiments to determine how species and

functional diversity loss impact decomposition rates in the prairie.

While it was a rare catch overall, no specimens of Staphylinus ornaticauda, a large

flightless, potentially vulnerable rove beetle, were found on burned sites, but were found on

multiple grazed and undisturbed sites. As well, a probably undescribed species of water skater,

Stenus sp. was found during this study, and was found on several sites, none of which were

burned. Burning the prairie likely removed the moss and litter layer necessary for water skater

115

development and for their prey. This may be a good example of the need to retain refugia in the

TGP, both from planned burns, and from wildfires when possible.

Though I did not detect spatial autocorrelation in any of my models, my sites were quite

clumped together out of time and travel necessity, as well as the limited size and disturbance

regimes of the preserve. Since I found a large number of new records for such a small area in the

province, it highlights the need to conduct further studies in other TGP remnants throughout the

province. To prevent destruction from animal or weather activity, I covered my pitfalls with a

rain cover. This probably had the undesired consequence of limiting the ability of the scent from

the baits to spread very far. As well, fish skin was responsible for most of the captures. Different

bait sources should be explored in subsequent studies to attract greater diversity of decomposers

and predators. It would also be interesting to compare bait efficiency of bison vs. cattle dung, as

native species may be more likely to respond positively to bison dung than that of cattle. Though

I performed a pilot test of measuring dung decomposition and comparing it to beetle diversity

(unpublished data), I ultimately decided to not report on it, as I determined that the size of the

dung pats I used were too small, and quickly desiccated before the decomposer fauna could

effectively start to displace it. If I or others were to try again, I would recommend large pats of

dung with regular measuring of water content. Dung beetles, and presumably other dung feeding

beetles, reduce greenhouse gas emissions, albeit very modestly in pastureland (Slade et al. 2016).

It is predicted that their contributions are increased when cattle, and by extension beetles, spend

more time out in a frost-free pasture, which could be predicted under increasing temperatures

expected with climate change. Detrital pathways are known to have significant effects on many

communities differing by ecological role in most food webs, not just directly on the decomposers

themselves (reviewed in Moore et al. 2004). Thus, it is imperative that abundance and diversity

of many trophic levels are related back to decomposition rates in the TGP to ensure proper

maintenance of its floral and faunal communities.

4.5 Conclusion

It is hoped that my study sparks an increase into the often-overlooked decomposer and

predatory insect communities in TGP and other grasslands. My study shows that though burning

and grazing have similar goals in shaping the prairie plant community, they have opposing

116

effects on beetle fauna. This can lead to deficits in the ecological functions performed by the

beetle community, including dung/carrion removal, pest fly suppression, moderation of

invertebrate herbivore populations, and nutrient cycling. Future studies should attempt to directly

measure the beneficial services of the decomposer beetle community. Since the TGP owes its

existence to both megafauna grazing and high intensity fires, it would be ideal for future studies

to examine the combined effect of both disturbance patterns on the prairie beetle fauna.

4.6 References

A. R. Westwood (pers. comm.). 2021. Personal Communication.

Adler, P.R., Sanderson, M.A., Weimer, P.J., and Vogel, K.P. 2009. Plant species composition and biofuel yields of conservation grasslands. Ecological Applications, 19: 2202–2209.

An, N., Price, K.P., and Blair, J.M. 2013. Estimating above-ground net primary productivity of the tallgrass prairie ecosystem of the Central Great Plains using AVHRR NDVI. International Journal of Remote Sensing, 34: 3717–3735. doi: 10.1080/01431161.2012.757376.

Anderson, D.G., and Nelson, T.D. 1993. United States patent number 5,229,126. Available from http://www.nrcresearchpress.com/doi/abs/10.1139/B08-083.

Anderson, R.S., and Peck, S.B. 1985. The insects and arachnids of Canada part 13: the carrion beetles of Canada and Alaska (Coleoptera: Silphidae and Agyrtidae). Canadian Government

Publishing Centre, Ottawa. p.

Baer, S.G., Kitchen, D.J., Blair, J.M., and Rice, C.W. 2002. Changes in ecosystem structure and function along a chronosequence of restored grasslands. Ecological Applications, 12: 1688–1701.

Baranowski, R. 1993. Revision of the genus Leiodes Latreille of North and Central America (Coleoptera: Leiodidae). Entomologica Scandinavica,: 5–149.

Barton, K. 2018. MuMIn: Multi-model inference.

Berger, H., Meyer, C.K., Mummert, A., Tirado, L., Saucedo, L., Bonello, H., Arsdale, D. Van,

and Williams, G. 2020. Land use dynamics within the tallgrass prairie ecosystem: the case for the Conservation Reserve Program ( CRP ). Theoretical Ecology, 13: 289–300. Available from https://doi.org/10.1007/s12080-020-00452-z.

Beucke, K. 2009. Beetles (Coleoptera : Geotrupidae , Scarabaeidae , and Silphidae) in Burned

and Unburned Pine Barrens in Long Island , New York. Entomologica Americana, 115: 168–171. doi: 10.2307/25622686.

Beutel, R.G., and Leschen, R.A.B. 2005. Coleoptera, beetles volume 1: morphology and systematics (Archostemata, Adephaga, Myxophaga, Polyphaga partim). Handbook of

Zoology, 4: 1–567.

Bohac, J. 1999. Staphylinid beetles as bioindicators. Agriculture, Ecosystems and Environment,

117

74: 357–372.

Bossenbroek, J.M., Wagner, H.H., and Wiens, J.A. 2004. Taxon-dependent scaling: beetles, birds, and vegetation at four North American grassland sites. Landscape Ecology, 20: 675–

688. doi: 10.1007/s10980-004-5651-4.

Bousquet, Y., Bouchard, P., Davies, A.E., and Sikes, D.S. 2013. Checklist of beetles (Coleoptera) of Canada and Alaska. Second edition. ZooKeys,: 1–402. doi: 10.3897/zookeys.360.4742.

Bousquet, Y., and Laplante, S. 2006. The insects and arachnids of Canada part 24: Coleoptera Histeridae. NRC Research Press, Ottawa. p.

Brunke, A., Newton, A., Klimaszewski, J., Majka, C., and Marshall, S. 2011. Staphylinidae of eastern Canada and adjacent United States. Key to subfamilies; Staphylininae: tribes and

subtribes, and species of Staphylinina. Canadian Journal of Arthropod Identification,: 1–110. doi: 10.3752/cjai.2011.12.

Brunke, A.J., Bahlai, C.A., Klimaszewski, J., and Hallett, R.H. 2014. Rove beetles (Coleoptera: Staphylinidae) in Ontario, Canada soybean agroecosystems: Assemblage diversity,

composition, seasonality, and habitat use. Canadian Entomologist, 146: 652–670. doi: 10.4039/tce.2014.19.

Bulan, C.A., and Barrett, G.W. 1971. The effects of two acute stresses on the arthropod component of an experimental grassland ecosystem. Ecology, 52: 597–605.

Buse, J., Šlachta, M., Sladecek, F.X.J., Pung, M., Wagner, T., and Entling, M.H. 2015. Relative importance of pasture size and grazing continuity for the long-term conservation of European dung beetles. Biological Conservation, 187: 112–119. doi: 10.1016/j.biocon.2015.04.011.

De Cáceres, M., and Jansen, F. 2016. indicspecies.

De Cáceres, M., and Legendre, P. 2009. Associations between species and groups of sites : indices and statistical inference. Ecology, 90: 3566–3574.

De Cáceres, M., Legendre, P., and Moretti, M. 2010. Improving indicator species analysis by

combining groups of sites. Oikos,: 1674–1684. doi: 10.1111/j.1600-0706.2010.18334.x.

Campbell, J.M. 1968. A revision of the new world Micropeplinae (Coleoptera: Staphylinidae) with a rearrangement of the world species. The Canadian Entomologist, 100: 225–267.

Campbell, J.M. 1979. A revision of the genus Tachyporus Gravenhorst (Coleoptera:

Staphylinidae) of North and Central America. Memoirs of the Entomological Society of Canada, 111: 1–95.

Campbell, J.M. 1983. A revision of the North American Omaliinae (coleoptera: Staphylinidae): The genus Olophrum. The Canadian Entomologist, 115: 577–622.

Campbell, J.M. 1984. A revision of the North American Omaliinae (Coleoptera: Staphylinidae): The genera Arpedium Erichson and Eucnecosum Reitter. The Canadian Entomologist, 116: 487–527.

118

Campbell, J.M. 1991. A revision of the genera Mycetoporus Mannerheim and Ischnosoma Stephens (Coleoptera: Staphylinidae: Tachyporinae) of North and Central America. Memoirs of the Entomological Society of Canada, 123: 1–169.

Campbell, J.M. 1993a. A revision of the genera Bryoporus Kraatz and Bryophacis Reitter and two new related genera from America north of Mexico (Coleoptera: Staphylinidae: Tachyporinae). Memoirs of the Entomological Society of Canada, 125.

Campbell, J.M. 1993b. A review of the species of Nitidotachinus new genus (Coleoptera:

Staphylinidae: Tachyporinae). The Canadian Entomologist, 125: 521–548.

Chao, A., Gotelli, N.J., Hsieh, T.C., Sander, E.L., and Colwell, R.K. 2014. Rarefaction and extrapolation with Hill numbers : a framework for sampling and estimation in species diversity studies. Ecological Monographs, 84: 45–67.

Cook, W.M., and Holt, R.D. 2006. Fire frequency and mosaic burning effects on a tallgrass prairie ground beetle assemblage. Biodiversity and Conservation, 15: 2301–2323. doi: 10.1007/s10531-004-8227-3.

Costanza, R., D’Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem,

S., O’Neill, R. V, Paruelo, J., Raskin, R.G., Sutton, P., and van den Belt, M. 1997. The value of the world’s ecosystem services and natural capital. Nature, 387: 253–260.

Crist, T.O., and Peters, V.E. 2014. Landscape and local controls of insect biodiversity in conservation grasslands: implications for the conservation of ecosystem service providers in

agricultural environments. Land, 3: 693–718. doi: 10.3390/land3030693.

Cruise, A., Watson, D.W., and Schal, C. 2018. Ecological succession of adult necrophilous insects on neonate Sus scrofa domesticus in central North Carolina. : 7–10.

Dennis, P., Aspinall, R.J., and Gordon, I.J. 2002. Spatial distribution of upland beetles in relation

to landform, vegetation and grazing management Basic and Applied Ecology. Basic Applied Ecology, 3: 183–193. doi: 10.1078/1439-1791-00081.

Dennis, P., Wratten, S.D., and Sotherton, W. 1990. Feeding behaviour of the staphylinid beetle Tachyporus hypnorum in relation to its potential for reducing aphid numbers in wheat.

Annals of Applied Biology, 117: 267–276.

Dennis, P., Young, M.R., and Gordon, I.J. 1998. Distribution and abundance of small insects and arachnids in relation to structural heterogeneity of grazed, indigenous grasslands. Ecological Entomology, 23: 253–264.

Doxon, E.D., Davis, C.A., Fuhlendorf, S.D., and Winter, S.L. 2011. Aboveground macroinvertebrate diversity and abundance in sand sagebrush prairie managed with the use of pyric herbivory. Rangeland Ecology & Management, 64: 394–403. doi: 10.2111/REM-D-10-00169.1.

Duffey, E. 1975. The effects of human trampling on the fauna of grassland litter. Biological Conservation, 7: 255–274.

Ehrenreich, J.H., and Aikman, J.M. 1963. An ecological study of the effect of certain management practices on native prairie in Iowa. Ecological Monographs, 33: 113–130.

119

Errouissi, F., and Jay-Robert, P. 2019. Consequences of habitat change in euromediterranean landscapes on the composition and diversity of dung beetle assemblages. Journal of Insect Conservation, 23: 15–28. doi: 10.1007/s10841-018-0110-8.

ESRI. 2018. ArcGIS Pro. Available from http://pro.arcgis.com.

Faber, S.M. 2008. Plant and soil feedback relationships in Manitoba tallgrass prairie. [degree] thesis, University of Manitoba, .

Farwig, N., Brandl, R., Siemann, S., Wiener, F., and Müller, J. 2014. Decomposition rate of

carrion is dependent on composition not abundance of the assemblages of insect scavengers. Oecologia, 175: 1291–1300. doi: 10.1007/s00442-014-2974-y.

Gerla, P.J., Cornett, M.W., Ekstein, J.D., and Ahlering, M.A. 2012. Talking big: Lessons learned from a 9000 hectare restoration in the northern tallgrass prairie. Sustainability, 4: 3066–

3087. doi: 10.3390/su4113066.

Good, J.A., and Giller, P.S. 1991. The diet of predatory staphylinid beetles - a review of records. Entomologist’s Monthly Magazine, 127: 77–89.

Gordon, R.D., and Skelley, P.E. 2007. A monograph of the Aphodiini inhabiting the United

States and Canada (Coleoptera: Scarabaeidae: Aphodiinae). Memoirs of the American Entomological Institute, 79: 1–580.

Grant, T.A., Madden, E.M., Shaffer, T.L., and Dockens, J.S. 2010. Effects of prescribed fire on vegetation and passerine birds in northern mixed-grass prairie. Journal of Wildlife

Management, 74: 1841–1851. doi: 10.2193/2010-006.

Griffiths, H.M., Louzada, J., Bardgett, R.D., Beiroz, W., França, F., Tregidgo, D., and Barlow, J. 2015. Biodiversity and environmental context predict dung beetle-mediated seed dispersal in a tropical forest field experiment. Ecology, 96: 1607–1619.

Guimarães, J.A., and Mendes, J. 1998. Succession and Abundance of Staphylinidae in Cattle Dung in Uberlândia , Brazil. Memorias do Instituto Oswaldo Cruz, 93: 1996–1999.

Halffter, G., and Edmonds, W.D. 1982. The nesting behavior of dung beetles (Scarabaeinae): an ecological and evolutive approach. Instituto de Ecologia, Mexico, D.F. p.

Hansen, A.K., Justesen, M.J., Kepfer-Rojas, S., Byriel, D.B., Pedersen, J., and Solodovnikov, A. 2018. Ecogeographic patterns in a mainland-island system in Northern Europe as inferred from the rove beetles (Coleoptera: Staphylinidae) on Læsø island. European Journal of Entomology, 115. doi: 10.14411/eje.2018.025.

Harvey, D.J., Hawes, C.J., Gange, A.C., Finch, P., Chesmore, D., and Farr, I. 2011. Development of non-invasive monitoring methods for larvae and adults of the stag beetle, Lucanus cervus. Insect Conservation and Diversity, 4: 4–14. doi: 10.1111/j.1752-4598.2009.00072.x.

Helgesen, R.G., and Post, R.L. 1967. Saprophagous Scarabaeidae (Coleoptera) of North Dakota. North Dakota Insects, p.

Hill, M.O. 1973. Diversity and evenness: a unifying notation and its consequences. Ecology, 54:

120

427–432.

von Hoermann, C., Jauch, D., Kubotsch, C., Reichel-Jung, K., Steiger, S., and Ayasse, M. 2018. Effects of abiotic environmental factors and land use on the diversity of carrion-visiting

silphid beetles (Coleoptera: Silphidae): a large scale carrion study. PLoS ONE, 13: e0196839. doi: 10.5061/dryad.r5c64.

Hofmann, T.A., and Mason, C.F. 2006. Importance of management on the distribution and abundance of Staphylinidae (Insecta: Coleoptera) on coastal grazing marshes. Agriculture,

Ecosystems and Environment, 114: 397–406. doi: 10.1016/j.agee.2005.12.001.

Hofstede, R.G.M. 1995. The effects of grazing and burning on soil and plant nutrient concentrations in Colombian paramo grasslands. Plant and Soil, 173: 111–132.

Hooper, D.U., Adair, E.C., Cardinale, B.J., Byrnes, J.E.K., Hungate, B.A., Matulich, K.L.,

Gonzalez, A., Duffy, J.E., Gamfeldt, L., and O’Connor, M.I. 2012. A global synthesis reveals biodiversity loss as a major driver of ecosystem change. Nature, 486: 105–108. doi: 10.1038/nature11118.

Hsieh, T.C., Ma, K.H., and Chao, A. 2019. iNEXT: Interpolation and extrapolation for species

diversity. Available from http://chao.stat.nthu.edu.tw/blog/software-download/.

Husson, F., Josse, J., Le, S., and Mazet, J. 2020. Multivariate exploratory data analysis and data mining. Available from http://factominer.free.fr.

Irmler, U., and Lipkow, E. 2018. Effect of environmental conditions on distribution patterns of

rove beetles. In Biology of rove beetles (Staphylinidae): life history, evolution, ecology and distribution. Edited by O. Betz, U. Irmler, and J. Klimaszewski. Springer, Cham, Switzerland. pp. 117–144.

Jakubec, P., and Růžička, J. 2015. Is the type of soil an important factor determining the local

abundance of carrion beetles (Coleoptera: Silphidae)? European Journal of Entomology, 112: 747–754. doi: 10.14411/eje.2015.071.

Jo, Y., and Smitley, D.R. 2003. Predation of Ataenius spretulus (Coleoptera: Scarabaeidae) eggs and grubs by species of Carabidae and Staphylinidae on golf courses in Michigan.

Environmental Entomology, 32: 1370–1376.

Jonas, J.L., Whiles, M.R., and Charlton, R.E. 2002. Aboveground invertebrate responses to land management differences in a Central Kansas grassland. Environmental Entomology, 31: 1142–1152.

Jost, L. 2007. Partitioning diversity into independent alpha and beta components. Ecology, 88: 2427–2439.

Joyce, J., and Morgan, J.P. 1989. Manitoba’s tall-grass prairie conservation project. Proceedings of the North American Prairie Conferences, 34: 71–74.

Klimaszewski, J., Brunke, A., Work, T.T., and Venier, L. 2018a. Rove beetles (Coleoptera, Staphylinidae) as bioindicators of change in boreal forests and their biological control services in agroecosystems: Canadian case studies. In Biology of rove beetles (Staphylinidae): life history, evolution, ecology and distribution. Edited by O. Betz, U.

121

Irmler, and J. Klimaszewski. Springer, Cham, Switzerland. pp. 161–181.

Klimaszewski, J., Godin, B., Davies, A., Bourdon, C., and Horwood, D. 2018b. Forty new records of aleocharine beetles, and two new species in the genera Acrotona Thomson and

Atheta Thomson, for the province of Manitoba, Canada (Coleoptera, Staphylinidae, Aleocharinae). Insecta Mundi,: 1–34.

van Klink, R., van der Plas, F., (Toos) van Noordwijk, C.G.E., WallisDeVries, M.F., and Olff, H. 2015. Effects of large herbivores on grassland arthropod diversity. Biological Reviews, 90:

347–366. doi: 10.1111/brv.12113.

Klink, R. Van, Rickert, C., Vermeulen, R., Vorst, O., Wallisdevries, M.F., and Bakker, J.P. 2013. Grazed vegetation mosaics do not maximize arthropod diversity : Evidence from salt marshes. Biological Conservation, 164: 150–157. doi: 10.1016/j.biocon.2013.04.023.

Kolbe, H.J. 1908. Mein system der coleopteren. Zeitschrift fur Wissenschaftliche Insektenbiologie, 4: 116–123.

Koper, N., Mozel, K.E., and Henderson, D.C. 2010. Recent declines in northern tall-grass prairies and effects of patch structure on community persistence. Biological Conservation,

143: 220–229. doi: 10.1016/j.biocon.2009.10.006.

Kral, K.C., Limb, R.F., Harmon, J.P., and Hovick, T.J. 2017. Arthropods and fire: previous research shaping future conservation. Rangeland Ecology & Management, 70: 589–598. doi: 10.1016/j.rama.2017.03.006.

Kruess, A., and Tscharntke, T. 2002. Contrasting responses of plant and insect diversity to variation in grazing intensity. Biological Conservation, 106: 293–302.

Langellotto, G.A., and Denno, R.F. 2004. Responses of invertebrate natural enemies to complex-structured habitats: a meta-analytical synthesis. Oecologia, 139: 1–10. doi: 10.1007/s00442-

004-1497-3.

Legendre, P., and Anderson, M.J. 1999. Distance-based redundancy analysis : Testing multispecies responses in ... Ecological Monographs, 69: 1–24.

Leivas, F., and Carneiro, E. 2012. Utilizando os hexápodes ( Arthropoda , Hexapoda ) como

bioindicadores na Biologia da Conservação : Avanços e perspectivas. Estudos de Biologia, 34: 203–213. doi: 10.7213/estud.biol.7333.

Liu, Y., Rothenwohrer, C., Scherber, C., Batary, P., Elek, Z., Steckel, J., Erasmi, S., Tscharntke, T., and Westphal, C. 2014. Functional beetle diversity in managed grasslands: effects of

region, landscape context and land use intensity. Landscape Ecology, 29: 529–540. doi: 10.1007/s10980-014-9987-0.

Ludecke, D., Makowski, D., Waggoner, P., and Patil, I. 2020. Assesment of regression models. Available from https://easystats.github.io/performance/.

Madsen, M., Overgaard Nielsen, B., Holter, P., Pedersen, O.C., Brochner Jespersen, J., Vagn Jensen, K.M., Nansen, P., and Gronvold, J. 1990. Treating cattle with Ivermectin: effects on the fauna and decomposition of dung pats. Journal of Applied Ecology, 27: 1–15.

122

Majka, C., and Klimaszewski, J. 2008. New records of Canadian Aleocharinae (Coleoptera: Staphylinidae). ZooKeys, 2: 85–114.

Marshall, S.A. 2018. Beetles: the Natural History and Diversity of Coleoptera. Firefly Books

Ltd., Richmond Hill. p.

McKenna, D.D., Farrell, B.D., Caterino, M.S., Farnum, C.W., Hawks, D.C., Maddison, D.R., Seago, A.E., Short, A.E.Z., Newton, A.F., and Thayer, M.K. 2015. Phylogeny and evolution of Staphyliniformia and Scarabaeiformia: forest litter as a stepping stone for diversification

of nonphytophagous beetles. Systematic Entomology, 40: 35–60. doi: 10.1111/syen.12093.

McLauchlan, K.K., Hobbie, S.E., and Post, W.M. 2006. Conversion from agriculture to grassland builds soil organic matter on decadal timescales. Ecological Applications, 16: 143–153.

Moore, J.C., Berlow, E.L., Coleman, D.C., de Ruiter, P.C., Dong, Q., Hastings, A., Collins Johnson, N., McCann, K.S., Melville, K., Morin, P.J., Nadelhoffer, K., Rosemond, A.D., Post, D.M., Sabo, J.L., Scow, K.M., Vanni, M.J., and Wall, D.H. 2004. Detritus, trophic dynamics and biodiversity. Ecology Letters, 7: 584–600. doi: 10.1111/j.1461-

0248.2004.00606.x.

Negro, M., Rolando, A., and Palestrini, C. 2011. The impact of overgrazing on dung beetle diversity in the Italian maritime alps. Environmental Entomology, 104: 1081–1092. doi: http://dx.doi.org/10.1603/EN11105.

Nelson, A.M., Moriasi, D.N., Fortuna, A.-M., Steiner, J.L., Starks, P.J., Northup, B., and Garbrecht, J. 2020. Runoff water quantity and quality data from native tallgrass prairie and crop–livestock systems in Oklahoma between 1977 and 1999. Journal of Environmental Quality, 49: 1062–1072. doi: 10.1002/jeq2.20075.

Newton, A., Thayer, M.K., Ashe, J.S., and Chandler, D.S. 2001. Staphylinidae Latrielle, 1802. In American Beetles vol. 1: Archostemata, Myxophaga, Adephaga, Polyphaga: Staphyliniformia. Edited by R.H. Arnett and M.C. Thomas. CRC Press, New York. pp. 272–418.

Nunes, C.A., Braga, R.F., de Moura Resende, F., de Siqueira Neves, F., Cortes Figueira, J.E., and Wilson Fernandes, G. 2018. Linking biodiversity, the environment and ecosystem functioning: ecological functions of dung beetles along a tropical elevational gradient. Ecosystems, 21: 1244–1254. doi: 10.1007/s10021-017-0216-y.

Oksanen, J., Blanchet, F.G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P., O’Hara, R.B., Simpson, G.L., Solymos, P., Stevens, M.H.H., Szoecs, E., and Wagner, H. 2018. vegan: community ecology package. Available from https://cran.r-project.org/package=vegan.

Orrock, J.L., Curler, G.R., Danielson, B.J., and Coyle, D.R. 2011. Large-scale experimental landscapes reveal distinctive effects of patch shape and connectivity on arthropod communities. Landscape Ecology, 26: 1361–1372. doi: 10.1007/s10980-011-9656-5.

Peck, S.B., and Cook, J. 2002. Systematics, distributions, and bionomics of the small carrion

beetles (Coleoptera: Leiodidae: Cholevinae: Cholevini) of North America. The Canadian

123

Entomologist, 134: 723–787.

Peck, S.B., and Cook, J. 2009. Review of the Sogdini of North and Central America (Coleoptera: Leiodidae: Leiodinae) with descriptions of fourteen new species and three new genera.

Zootaxa, 2102: 1–74.

Peck, S.B., and Cook, J. 2013. Systematics and distributions of the genera Cyrtusa Erichson, Ecarinosphaerula Hatch, Isoplastus Horn, Liocyrtusa Daffner, Lionothus Brown, and Zeadolopus Broun of the United States and Canada (Coleoptera: Leiodid. Insecta Mundi,

0310: 1–32.

Peck, S.B., and Stephan, K. 1996. A revision of the genus Colon Herbst (Coleoptera; Leiodidae; Coloninae) of North America. The Canadian Entomologist, 128: 667–741.

Pyne, S.J. 1986. “These conflagrated prairies” a cultural fire history of the grasslands. In The

Prairie: past, present, and future: Proceedings of the ninth North American prairie conference part 6. Fire and prairie ecosystems. Edited by G.K. Clambey and R.H. Pemble.

R Core Team. 2017. R: a language and environment for statistical computing. Available from https://www.r-project.org.

Reed, H.E., Blair, J.M., Wall, D.H., and Seastedt, T.R. 2009. Impacts of management legacies on litter decomposition in response to reduced precipitation in a tallgrass prairie. Applied Soil Ecology, 42: 79–85. doi: 10.1016/j.apsoil.2009.01.009.

Rényi, A. 1961. On measures of entropy and information. In 4th Berkeley symposium on

mathematical statistics and probability. Edited by J. Neyman.

Reynolds, C., Fletcher, R.J., Carneiro, C.M., Jennings, N., Ke, A., LaScaleia, M.C., Lukhele, M.B., Mamba, M.L., Sibiya, M.D., Austin, J.D., Magagula, C.N., Mahlaba, T., Monadjem, A., Wisely, S.M., and McCleery, R.A. 2018. Inconsistent effects of landscape heterogeneity

and land-use on animal diversity in an agricultural mosaic: a multi-scale and multi-taxon investigation. Landscape Ecology, 33: 241–255. doi: 10.1007/s10980-017-0595-7.

Rocha-Ortega, M., and Coronel-Arellano, H. 2019. How predictable are the responses of ant and dung beetle assemblages to patch and landscape attributes in fragmented tropical forest

landscapes? Landscape and Ecological Engineering, 15: 315–322. doi: 10.1007/s11355-018-0367-9.

Rös, M., Escobar, F., and Halffter, G. 2012. How dung beetles respond to a human-modified variegated landscape in Mexican cloud forest: a study of biodiversity integrating ecological

and biogeographical perspectives. Diversity and Distributions, 18: 377–389. doi: 10.1111/j.1472-4642.2011.00834.x.

Roughley, R.E., Pollock, D.A., and Wade, D.J. 2006. Biodiversity of ground beetles (Coleoptera: Carabidae) and spiders (Araneae) across a tallgrass prairie – aspen forest ecotone in

southern Manitoba. The Canadian Entomologist, 138: 545–567.

Roughley, R.E., Pollock, D.A., and Wade, D.J. 2010. Tallgrass prairie, ground beetles (Coleoptera: Carabidae), and the use of fire as a biodiversity and conservation management tool. In Arthropods of Canadian Grasslands (Volume 1): Ecology and Interactions in

Grassland Habitats. pp. 227–235. doi: 10.3752/9780968932148.ch10.

124

Royston, P. 1982a. An extension of Shapiro and Wilk’s test for normality to large samples. Applied Statistics, 31: 115–124. doi: 10.2307/2347973.

Royston, P. 1982b. Algorithm AS 181: the W test for normality. Applied Statistics, 31: 176–180.

doi: 10.2307/2347986.

Royston, P. 1995. Remark AS R94: a remark on algorithm AS 181: the W test for normality. Applied Statistics, 44: 547–551. doi: 10.2307/2986146.

Sakamoto, Y., Ishiguro, M., and Kitagawa, G. 1986. Akaike information criterion statistics. D.

Reidel Publishing Company, p.

Samson, F., and Knopf, F. 1994. Prairie conservation in North America. BioScience, 44: 418–421.

Sánchez-de-Jesús, H.A., Arroyo-Rodríguez, V., Andresen, E., and Escobar, F. 2016. Forest loss

and matrix composition are the major drivers shaping dung beetle assemblages in a fragmented rainforest. Landscape Ecology, 31: 843–854. doi: 10.1007/s10980-015-0293-2.

Sands, B.O., and Wall, R. 2017. Dung beetles reduce livestock gastrointestinal parasite availability on pasture. Journal of Applied Ecology, 54: 1180–1189. doi: 10.1111/1365-

2664.12821/abstract.

Seastedt, T.R. 1984. Microarthropods of burned and unburned tallgrass prairie. Journal of the Kansas Entomological Society, 57: 468–476.

Seastedt, T.R., Hayes, D.C., and Petersen, N.J. 1986. Effects of vegetation, burning, and mowing

on soil macroarthropods of tallgrass prairie. In The prairie: past, present, and future : proceedings of the ninth North American prairie conference. Edited by G.K.. Clambey and R.. Pemble.

Short, A.E.Z., and Fikacek, M. 2013. Molecular phylogeny, evolution and classification of the

Hydrophilidae (Coleoptera). Systematic Entomology, 38: 723–752. doi: 10.1111/syen.12024.

Siemann, E., Tilman, D., Haarstad, J., and Ritchie, M. 1998. Experimental tests of the dependence of arthropod diversity on plant diversity. The American Naturalist, 152: 738–

750.

Slade, E.M., Riutta, T., Roslin, T., and Tuomisto, H.L. 2016. The role of dung beetles in reducing greenhouse gas emissions from cattle farming. Scientific Reports, 6: 1–9. doi: 10.1038/srep18140.

Smetana, A. 1971. Revision of the tribe Quediini of America north of Mexico (Coleoptera: Staphylinidae). Memoirs of the Entomological Society of Canada, 103: 1–303.

Smetana, A. 1988. Review of the family Hydrophilidae of Canada and Alaska. Memoirs of the Entomological Society of Canada, 120: 3–316.

Smetana, A. 1995. Rove beetles of the subtribe Philonthina of America north of Mexico (Coleoptera: Staphylinidae). Memoirs on Entomology, International, 3: 1–945.

Somay, L., Szigeti, V., Boros, G., Ádám, R., and Báldi, A. 2021. Wood pastures: a transitiona l

125

habitat between forests and pastures for dung beetle assemblages. Forests, 12: 1–16.

Stoner, K.J.L., and Joern, A. 2004. Landscape vs. local habitat scale influences to insect communities from tallgrass prairie remnants. Ecological Applications, 14: 1306–1320.

Sveinson, J.M.A. 2003. Restoring tallgrass prairie in southern Manitoba, Canada. [degree] thesis, University of Manitoba, .

Swengel, A.B. 2001. A literature review of insect responses to fire, compared to other conservation managements of open habitat. Biodiversity and Conservation, 10: 1141–1169.

Taft, J.B., and Kron, Z.P. 2014. Evidence of species and functional group attrition in shrub -encroached prairie: implications for restoration. American Midland Naturalist, 172: 252–265.

Tester, J.R., and Marshall, W.H. 1961. A study of certain plant and animal interrelations on a

native prairie in northwestern Minnesota. Occasional Papers of the Minnesota Museum of Natural History, 8.

Thayer, M.K. 2016. 14.7 Staphylinidae. In Handbook of Zoology Volume 1, Second Edi. Edited by R.G. Beutel and R.A.B. Leschen. De Gruyter, Jena. pp. 394–442.

The Biological Survey of Canada. 1994. Terrestrial arthropod biodiversity: planning a study and recommended sampling techniques.

Tocco, C., Negro, M., Rolando, A., and Palestrini, C. 2013. Does natural reforestation represent a potential threat to dung beetle diversity in the Alps? Journal of Insect Conservation, 17:

207–217. doi: 10.1007/s10841-012-9498-8.

Tomer, M.D., Schilling, K.E., and Cole, K.J. 2019. Nitrate on a slow decline: Watershed water quality response during two decades of tallgrass prairie ecosystem reconstruction in Iowa. Journal of Environmental Quality, 48: 579–585. doi: 10.2134/jeq2018.07.0258.

Tonelli, M., Verdú, J.R., Morelli, F., and Zunino, M. 2020. Dung beetles: functional identity, not functional diversity, accounts for ecological process disruption caused by the use of veterinary medical products. Journal of Insect Conservation, 24: 643–654. doi: 10.1007/s10841-020-00240-4.

Tóthmérész, B., Nagy, D.D., Mizser, S., Bogyó, D., and Magura, T. 2014. Edge effects on ground-dwelling beetles (Carabidae and Staphylinidae) in oak forest-forest edge-grassland habitats in Hungary. European Journal of Entomology, 111: 686–691. doi: 10.14411/eje.2014.091.

Veech, J.A. 2006. A comparison of landscapes occupied by increasing and decreasing populations of grassland birds. Conservation Biology, 20: 1422–1432. doi: 10.1111/j.1523-1739.2006.00487.x.

Verdú, J.R., Numa, C., and Hernández-Cuba, O. 2011. The influence of landscape structure on

ants and dung beetles diversity in a Mediterranean savanna — forest ecosystem. Ecological Indicators, 11: 831–839. doi: 10.1016/j.ecolind.2010.10.011.

Vieira, L., Nascimento, P.K.S., and Leivas, F.W.T. 2018. Habitat association promotes diversity

126

of histerid beetles (Coleoptera: Histeridae) in neotropical ecosystems. 72: 541–549.

Wade, D.J. 2002. The effect of burn season on the spider (Araneae) fauna of tallgrass prairie and its implications on prairie management. [degree] thesis, University of Manitoba, .

Walker Jr, T.L., and Hoback, W.W. 2007. Effects of invasive Eastern Redcedar on capture rates of Nicrophorus americanus and other Silphidae. Environmental Entomology, 36: 297–307.

Wall, R., and Strong, L. 1987. Environmental consequences of treating cattle with the antiparasitic drug Ivermectin. Nature, 327: 418–421.

Watrous, L. 1980. Lathrobium (Tetartopeus): natural history, phylogeny and revision of the nearctic species (Coleoptera, Staphylinidae). Systematic Entomology, 5: 303–338.

Weithmann, S., Kuppler, J., Degasperi, G., Steiger, S., Ayasse, M., and Von Hoermann, C. 2020. Rove beetle (Coleoptera: Staphylinidae) communities in German forests. Insects, 11: 1–16.

Westwood, R., Westwood, A.R., Hooshmandi, M., Pearson, K., Lafrance, K., and Murray, C. 2020. A field-validated species distribution model to support management of the critically endangered Poweshiek skipperling (Oarisma poweshiek) butterfly in Canada. Conservation Science and Practice, 2: e163. doi: 10.1111/csp2.163.

Whiles, M.R., and Charlton, R.E. 2006. The ecological significance of tallgrass prairie arthropods. Annual Review of Entomology, 51: 387–412. doi: 10.1146/annurev.ento.51.110104.151136.

White, R.E. 1998. The Beetles of North America. Edited ByR.T. Peterson. Houghton Mifflin

Harcourt, Boston. p.

Wu, X., Duffy, J.E., Reich, P.B., and Sun, S. 2011. A brown-world cascade in the dung decomposer food web of an alpine meadow: effects of predator interactions and warming. Ecological Monographs, 81: 313–328.

Zavaleta, E.S., Pasari, J.R., Hulvey, K.B., and Tilman, G.D. 2010. Sustaining multiple ecosystem functions in grassland communities requires higher biodiversity. PNAS, 107: 1443–1446. doi: 10.1073/pnas.0906829107.

127

4.7 Chapter 4 Supplementary Materials

Figure 4.9 Graph of the first two axes of a principal coordinate analysis performed on percent

cover on the various landscapes found at 500 m from the centre of study sites.

128

Figure 4.10 Graph showing the first two axes of a PCA performed on groundcover data obtained

by averaging two rounds of quadrat sampling performed in 2018. 16 quadrats were recorded

from each study site and then averaged by site. This was done twice, once in June, and once in

August. The site averages of the two rounds were once again averaged to obtain the results put in

to the PCA.

129

Haplogastran Beetles Model Tables

Table 4.5 Results of GLMMs performed on haplogastran beetle pitfall trap capture abundance,

species richness, and Hill’s Shannon Effective Species Number. Site was used as a random effect in all models. The models represent the submodel with the lowest AICc score.

Response

Variable Fixed Effects Est. Std. Error z value p value

Abundance

Intercept 7.551 0.602 12.55 <0.001

Year (2017) -3.08 0.111 -27.83 <0.001

Day-of-year -0.01 0.003 -2.92 0.004

Avg. Precip. -0.10 0.041 -2.53 0.012

Landcover PCA @ 500 m 0.123 0.038 3.23 0.001

Treatment: Grazed 0.483 0.158 3.06 0.002

Treatment: Undisturbed -0.20 0.164 -1.24 0.215

Species

Richness

Intercept 4.635 0.286 16.19 <0.001

Year (2017) -1.33 0.067 -19.78 <0.001

Day-of-year -0.01 0.001 -5.04 <0.001

Avg. Precip. -0.05 0.021 -2.26 0.024

Landcover PCA @ 500 m 0.037 0.016 2.30 0.021

Groundcover PCA -0.03 0.017 -1.69 0.091

Hill’s

Shannon

Effective

Species

Number

Intercept 1.932 0.085 22.66 <0.001

Year (2017) -0.27 0.062 -4.37 <0.001

Landcover PCA @ 500 m 0.044 0.026 1.70 0.090

Treatment: Grazed -0.36 0.107 -3.38 0.001

Treatment: Undisturbed -0.21 0.108 -1.96 0.050

130

Model Distribution Dispersion Parameter

Random Effect S.D.

AIC Log

Likelihood R2C

Abundance Negative

Binomial 4.21 (+/- 0.70) 0.097 1079.7 -530.843 0.463

Sp. Rich. Poisson N/A 0.043 544.4 -265.2 0.874

Hill’s

Shannon

Gamma

(Log Link) 9.99 (+/- 1.47) 0.096 417 -201.48 0.318

Table 4.6 Results of GLMMs performed on predatory haplogastran beetle pitfall trap capture abundance, species richness, and Hill’s Shannon Effective Species Number. Site was used as a

random effect in all models. The models represent the submodel with the lowest AICc score.

Response

Variable Fixed Effects Est. Std. Error z value p value

Abundance

Intercept 8.060 0.596 13.51 <0.001

Year (2017) -1.66 0.118 -14.06 <0.001

Day-of-year -0.02 0.003 -7.77 <0.001

Avg. Precip. -0.12 0.043 -2.86 0.004

Treatment: Grazed -0.38 0.149 -2.54 0.011

Treatment: Undisturbed -0.47 0.145 -3.26 0.001

Groundcover PCA -0.08 0.036 -2.30 0.021

Species

Richness

Intercept 23.942 2.163 11.07 <0.001

Year (2017) -5.56 0.417 -13.32 <0.001

Day-of-year -0.07 0.010 -7.00 <0.001

Avg. Precip. -0.50 0.159 -3.15 0.002

Groundcover PCA -0.40 0.177 -2.26 0.024

Hill’s

Shannon

Effective

Species

Number

Intercept 16.086 1.718 9.36 <0.001

Year (2017) -3.55 0.331 -10.74 <0.001

Day-of-year -0.05 0.008 -5.63 <0.001

Avg. Precip. -0.26 0.126 -2.09 0.037

131

Groundcover PCA -0.23 0.149 -1.56 0.118

Model Distribution Dispersion

Parameter/Residual Variance

Random Effect S.D.

AIC Log

Likelihood R2C

Abundance Negative Binomial

6.88 (+/- 2.02) 0.117 631.9 -306.948 0.208

Sp. Rich. Gaussian 1.99 (+/- 0.15) 0.891 483.3 -234.667 0.744

Hill’s Shannon

Gaussian 1.58 (+/- 0.12) 0.787 435.4 -210.723 0.668

Table 4.7 Results of GLMMs performed on decomposer haplogastran beetle pitfall trap capture abundance, species richness, and Hill’s Shannon Effective Species Number. Site was used as a random effect in all models. The models represent the submodel with the lowest AICc score.

Response

Variable Fixed Effects Est. Std. Error z value p value

Abundance

Intercept 6.220 0.725 8.58 <0.001

Year (2017) -3.41 0.138 -24.69 <0.001

Avg. Temp -0.09 0.036 -2.36 0.018

Avg. Precip. -0.13 0.053 -2.44 0.015

Landcover PCA @ 500 m 0.307 0.048 6.37 <0.001

Landscape MSI @ 500 m 0.124 0.077 1.62 0.106

Treatment: Grazed 0.369 0.197 1.88 0.061

Treatment: Undisturbed -0.33 0.189 -1.72 0.085

Species

Richness

Intercept 22.029 1.964 11.22 <0.001

Year (2017) -10.85 0.383 -28.35 <0.001

Day-of-year -0.04 0.009 -4.32 <0.001

Avg. Precip. 0.268 0.144 -1.87 0.062

Landcover PCA @ 500 m 0.463 0.106 4.38 <0.001

Hill’s

Shannon

Effective

Intercept 6.63 2.936 2.26 0.024

Year (2017) -4.75 0.303 -15.67 <0.001

132

Species

Number

Day-of-year -0.01 0.009 -1.68 0.094

Avg. Temp. 0.183 0.088 2.08 0.037

Model Distribution Dispersion

Parameter/Residual Variance

Random Effect S.D.

AIC Log

Likelihood R2C

Abundance Negative

Binomial 2.85 (+/- 0.49) <0.001 890.9 -435.452 0.574

Sp. Rich. Gaussian 1.83 (+/- 0.14) 0.168 452.3 -219.144 0.903

Hill’s

Shannon Gaussian 1.54 (+/- 0.10) <0.001 412.1 -200.053 0.731

Table 4.8 Results of a GLMM performed on aleocharine beetle abundance. Site was used as a

random effect. The model represents the submodel with the lowest AICc score.

Response Variable

Fixed Effects

Estimate Std. Error z value p value

Abundance

Intercept 6.881 0.826 8.33 <0.001

Day of Year -0.01 0.004 -2.78 0.005

Year: 2017 -3.27 0.153 -21.41 <0.001

Treatment: Grazed

1.172 0.258 4.55 <0.001

Treatment: Undisturbed

0.175 0.252 0.70 0.486

Model Distribution Dispersion

Parameter/Residual

Variance

Random Effect

S.D.

AIC Log

Likelihood R2C

Abundance Negative Binomial

2.34 (+/- 0.44) 0.304 962.9 -474.431 0.632

Because of the amount of data required to run the ‘Indval.g’ command and allow species to be

indicators of multiple sites, the ‘duleg=T’ function was added to allow a species to only be an

indicator for one site. The ‘r.g.’ method, however, allowed for species to be indicators of

multiple combinations of sites.

133

Table 4.9 Results of Indicator Species Analysis on beetle species by site using the Indval.g method. Guild refers to feeding guild that beetle belongs to, with P = predator, O = omnivore, and D = decomposer. Analysis was run for 999 permutations. Only species caught 10 or more

times are included.

Species Guild # Caught Site Method Statistic p-value

Tachyporus canadensis

O 22 Clough Indval.g 0.477 0.039

Nitidotachinus tachyporoides

O 19 Forrester &

Casson Indval.g 0.530 0.004

Staphylinus ornaticauda

P 13 Forrester &

Casson Indval.g 0.519 0.009

Nicrophorus hebes D 1,044 Nickel Indval.g 0.412 0.049

Neobisnius jucundus P 74 Saranchuk Indval.g 0.458 0.042

Apocellus sp. P 98 Singh NW Indval.g 0.707 0.001

Heterosilpha ramosa

D 261 Singh NW Indval.g 0.621 0.006

Philonthus cognatus P 26 Smook SW Indval.g 0.453 0.039

Catops alsiosus D 1,451 Stefanchuk

2 Indval.g 0.456 0.034

Anotylus niger P 72 Storoschuk Indval.g 0.509 0.021

Table 4.10 Results of Indicator Species Analysis on beetle species by site using the r.g. method.

Guild refers to feeding guild that beetle belongs to, with P = predator, O = omnivore, and D = decomposer. Analysis was run for 999 permutations. Only species caught 10 or more times are included.

Species Guild # Caught Site(s) Method Statistic p-value

Nitidotachinus tachyporoides

O 19 Kai Kwong

Lee + Forrester & Casson

r.g. 0.465 0.033

Staphylinus

ornaticauda P 13

Forrester & Casson +

Stefanchuk r.g. 0.462 0.05

Stenus new sp. P 10 Smook NW + Stefanchuk +

Townsend r.g. 0.538 0.008

Apocellus sp. P 98

Dykun + Fisher +

Saranchuk + Sing

h NW + Storoschuk

r.g. 0.542 0.001

Neobisnius jucundus P 74 Fisher + Penner + Rose + Saranchuk

+ Singh NW r.g. 0.636 0.001

Anotylus niger P 72

Bott + Dykun +

Fisher + Fisher 2

+ Penner + Storoschuk + Townsend

r.g. 0.472 0.02

Heterosilpha D 261 Bott + Dykun + r.g. 0.446 0.033

134

ramosa Fisher + Fisher 2

+

Penner + Rose + Singh NW + Smo

ok NW + Smook

SW + Storoschuk

Table 4.11 List of species caught in pitfall traps in 2017 along with their feeding guild (P =

predator, D = decomposer, O = omnivore, N/A = unknown) and distribution of numbers caught

between the three disturbance types: B = burned, G = grazed, U = undisturbed. The other

columns represent the types of baits they were caught in, with C.D. = cow dung, A.A. = EtOH +

amyl acetate, F.S. = fish skin, and G.R. = ginger root.

Species Author Family Role B G U C.D. A.A. F.S. G.R. Total

Aleocharinae spp. Fleming 1821 Staph. N/A 52 227 91 89 83 135 63 370

Anacaena limbata (Fabricius 1792) Hydro. D 0 1 1 0 0 0 2 2

Anotylus niger (LeConte 1877) Staph. P 1 5 0 2 2 2 0 6

Aphodius fimetarius (Linnaeus 1758) Scarab. D 0 0 1 1 0 0 0 1

Apocellus sphaericollis (Say 1831) Staph. P 14 4 3 7 2 8 4 21

Baeocera sp. Erichson 1845 Staph. D 1 0 0 0 0 0 1 1

Bledius strenuus Casey 1889 Staph. D 2 0 0 1 1 0 0 2

Bryoporus rufescens LeConte 1863 Staph. N/A 0 0 2 1 0 1 0 2

Catops alsiosus (Horn 1885) Leiod. D 3 44 12 5 10 44 0 59

Colon tibiale Hatch 1957 Leiod. D 0 3 1 1 0 3 0 4

Creophilus maxillosus (Linnaeus 1758) Staph. P 1 0 0 0 0 1 0 1

Cymbiodyta minima Notman 1919 Hydro. D 0 0 2 0 1 0 1 2

Diapterna hyperborea (LeConte 1850) Scarab. D 3 1 2 2 0 3 1 6

Diapterna pinguella (Brown 1929) Scarab. D 4 13 28 1 38 3 3 45

Dinothenarus badipes (LeConte 1863) Staph. P 2 1 1 1 1 2 0 4

Erichsonius pusio (Horn 1884) Staph. P 0 1 0 0 0 1 0 1

Euspilotus assimilis (Paykull 1811) Hister. P 0 1 0 0 0 1 0 1

Gabrius nigritulus (Gravenhorst 1802) Staph. P 0 1 0 0 0 1 0 1

Gabrius picipennis (Mäklin 1852) Staph. P 1 2 1 1 0 1 2 4

Helophorus angusticollis

Orchymont 1945 Hydro. D 2 1 0 0 2 1 0 3

Heterosilpha ramosa (Say 1823) Silph. D 3 2 1 3 0 2 1 6

Hister paykullii Kirby 1837 Hister. P 1 0 1 0 2 0 0 2

Hydnobius substriatus LeConte 1863 Leiod. D 1 0 1 0 0 1 1 2

Ischnosoma undescribed species

Stephens 1829 Staph. N/A 0 0 1 0 0 0 1 1

Ischnosoma pictum (Horn 1877) Staph. N/A 2 1 4 1 2 4 0 7

Lathrobium sparsellum

Casey 1905 Staph. P 0 0 1 0 0 1 0 1

Leiodes assimilis (LeConte 1850) Leiod. D 6 3 7 2 2 10 2 16

Mycetoporus lucidulus LeConte 1863 Staph. N/A 0 0 2 0 0 0 2 2

Necrophila americana (Linnaeus 1758) Silph. D 1 23 6 3 4 23 0 30

Neobisnius jucundus (Horn 1884) Staph. P 0 3 0 0 2 0 1 3

Nicrophorus hebes Kirby 1837 Silph. D 14 43 19 9 36 30 1 76

Nicrophorus orbicollis Say 1825 Silph. D 0 3 1 0 0 3 1 4

Nicrophorus tomentosus

Weber 1801 Silph. D 1 1 0 0 0 2 0 2

Nitidotachinus tachyporoides

(Horn 1877) Staph. O 2 0 2 0 1 1 2 4

135

Oiceoptoma noveboracense

(Forster 1771) Silph. D 0 9 0 0 4 5 0 9

Olophrum rotundicolle Erichson 1839 Staph. P 1 0 0 0 1 0 0 1

Ontholestes cingulatus (Gravenhorst 1802) Staph. P 3 0 1 1 1 0 2 4

Onthophagus hecate (Panzer 1794) Scarab. D 0 0 3 2 0 1 0 3

Onthophagus nuchicornis

(Linnaeus 1758) Scarab. D 1 0 0 0 1 0 0 1

Pachystilicus hanhami (Wickham 1898) Staph. P 0 0 1 1 0 0 0 1

Paederus littorarius Gravenhorst 1806 Staph. P 10 5 8 7 4 7 5 23

Philonthus blossius Smetana 1995 Staph. P 1 0 0 0 0 0 1 1

Philonthus carbonarius

(Gravenhorst 1802) Staph. P 0 1 0 0 1 0 0 1

Philonthus cognatus (Stephens 1832) Staph. P 5 3 3 1 6 2 2 11

Philonthus couleensis Hatch 1957 Staph. P 1 0 0 0 1 0 0 1

Philonthus cruentatus (Gmelin 1790) Staph. P 0 1 1 2 0 0 0 2

Philonthus lomatus

Species Group

Erichson 1840 Staph. P 0 0 1 0 0 0 1 1

Philonthus nigrita Species Group

(Gravenhorst 1806) Staph. P 17 14 18 12 9 20 8 49

Philonthus schwarzi Horn 1884 Staph. P 16 5 4 5 3 11 6 25

Philonthus varians (Paykull 1789) Staph. P 8 3 2 7 4 0 2 13

Staphylinus ornaticauda

LeConte 1863 Staph. P 0 1 1 2 0 0 0 2

Stenus melanarius Stephens 1833 Staph. P 0 0 1 0 0 1 0 1

Stenus undescribed species

Latreille 1796 Staph. P 0 1 2 1 1 1 0 3

Tetartopeus furvulus Casey 1905 Staph. P 0 0 1 0 0 1 0 1

Thanatophilus lapponicus

(Herbst 1793) Silph. D 0 1 1 2 0 0 0 2

Tympanophorus puncticollis

(Erichson 1840) Staph. P 1 1 4 0 1 3 2 6

Total = 56 species Total

181 429 243 173 226 336 118 853 Avg. Total per site

36.2 72 34.7

Number of Species

32 34 40

Avg. Number of Species per site

6.4 5.7 5.71

Table 4.12 List of species caught in pitfall traps in 2018 along with their feeding guild (P =

predator, D = decomposer, O = omnivore, H = herbivore, N/A = unknown) and distribution of

numbers caught between the three disturbance types: B = burned, G = grazed, U = undisturbed.

The other columns represent the types of baits they were caught in, with C.D. = cow dung, A.A.

= EtOH + amyl acetate, F.S. = fish skin, and G.R. = ginger root.

Species Author Family Role B G U C.D. A.A. F.S. G.R. Total

Aleocharinae spp. Fleming 1821 Staph. N/A 1651 5407 3055 1262 1014 5968 1869 10113

Acrotrichis sp. Motschulsky 1848 Ptil. D 0 0 1 0 1 0 0 1

Acylophorus pronus Erichson 1840 Staph. P 1 0 0 1 0 0 0 1

Anacaena limbata (Fabricius 1792) Hydro. D 2 6 11 3 7 7 2 19

Anotylus niger (LeConte 1877) Staph. P 8 49 9 16 22 18 10 66

Apocellus sphaericollis

(Say 1831) Staph. P 46 15 16 13 16 32 16 77

Arpedium brunnescens

J.R. Sahlberg 1871 Staph. P 0 0 1 0 0 0 1 1

Ataenius spretulus (Haldeman 1848) Scarab. D 0 0 1 1 0 0 0 1

Atholus falli (Bickhardt 1912) Hister. P 2 0 0 2 0 0 0 2

Bisnius instabilis (Horn 1884) Staph. P 0 1 0 0 0 1 0 1

136

Bledius strenuus Casey 1889 Staph. D 13 1 3 1 5 5 6 17

Bolitobius sp. 1 Leach 1819 Staph. N/A 0 0 1 0 0 0 1 1

Bolitobius sp. 2 Leach 1819 Staph. N/A 0 0 3 3 0 0 0 3

Bryoporus rufescens LeConte 1863 Staph. N/A 0 2 4 1 2 1 2 6

Carpelimus sp. 1 Leach 1819 Staph. D 0 1 0 0 0 1 0 1

Carpelimus sp. 2 Leach 1819 Staph. D 1 0 1 0 1 0 1 2

Catops alsiosus (Horn 1885) Leiod. D 237 520 635 9 1 1312 70 1392

Catops gratiosus (Blanchard 1915) Leiod. D 2 9 5 0 0 15 1 16

Cercyon roseni Knish 1922 Hydro. D 4 7 10 3 3 9 6 21

Colobopterus

erraticus

(Linnaeus 1758) Scarab. D 1 0 0 1 0 0 0 1

Colon near politum Peck and Stephan

1996

Leiod. D 2 9 2 2 10 0 1 13

Creophilus maxillosus (Linnaeus 1758) Staph. P 1 1 2 0 1 3 0 4

Cymbiodyta minima Notman 1919 Hydro. D 0 0 1 0 0 0 1 1

Cymbiodyta vindicata Fall 1924 Hydro. D 0 0 1 0 1 0 0 1

Cyrtusa subtestacea (Gyllenhal 1813) Leiod. D 0 1 1 2 0 0 0 2

Decarthron abnorme (LeConte 1849) Staph. P 0 1 1 1 0 1 0 2

Diapterna hyperborea (LeConte 1850) Scarab. D 11 5 19 11 6 11 7 35

Diapterna pinguella (W.J. Brown 1929) Scarab. D 44 145 170 24 290 33 12 359

Dinothenarus badipes (LeConte 1863) Staph. P 26 67 78 20 96 27 28 171

Erichsonius pusio (Horn 1884) Staph. P 2 1 0 1 0 2 0 3

Euconnus flavitarsis (LeConte 1852) Staph. P 1 0 2 0 1 1 1 3

Euspilotus assimilis (Paykull 1811) Hister. P 49 57 57 0 0 163 0 163

Gabrius picipennis (Mäklin 1852) Staph. P 18 12 16 13 8 12 13 46

Helophorus

angusticollis

Orchymont 1945 Hydro. D 3 6 4 5 3 2 3 13

Helophorus lacustris LeConte 1850 Hydro. D 1 0 0 0 0 1 0 1

Heterosilpha ramosa (Say 1823) Silph. D 171 68 16 65 52 118 20 255

Hister abbreviatus Fabricius 1775 Hister. P 3 1 2 0 0 6 0 6

Hister furtivus LeConte 1860 Hister. P 2 2 4 0 1 7 0 8

Hister paykullii Kirby 1837 Hister. P 1 3 3 0 1 5 1 7

Hydnobius substriatus

LeConte 1863 Leiod. D 0 0 1 0 1 0 0 1

Hydrochara obtusata (Say 1823) Hydro. D 0 1 0 0 1 0 0 1

Hydrochus granulatus Blatchley 1910 Hydro. D 1 0 0 0 1 0 0 1

Hydrochus neosquamifer

Smetana 1988 Hydro. D 0 1 7 0 0 7 1 8

Ischnosoma undescribed species

Stephens 1829 Staph. N/A 0 0 4 1 2 1 0 4

Ischnosoma pictum (Horn 1877) Staph. N/A 3 7 10 6 5 7 2 20

Lathrobium sparsellum

Casey 1905 Staph. P 0 1 1 0 1 0 1 2

Lathrobium sp. 1 Gravenhorst 1802 Staph. P 0 0 2 0 1 1 0 2

Lathrobium sp. 2 Gravenhorst 1802 Staph. P 1 0 1 0 0 1 1 2

Leiodes neglecta Baranowski 1993 Leiod. D 42 39 121 31 104 41 26 202

Margarinotus immunis

(Erichson 1834) Hister. P 0 1 0 1 0 0 0 1

Micropeplus sculptus LeConte 1863 Staph. D 0 4 0 0 1 1 2 4

Mycetoporus lucidulus

LeConte 1863 Staph. N/A 1 1 0 0 1 0 1 2

Necrophila americana (Linnaeus 1758) Silph. D 205 1142 933 2 3 2248 27 2280

Neobisnius jucundus (Horn 1884) Staph. P 53 16 2 16 23 15 17 71

Neohypnus obscurus (Erichson 1839) Staph. P 2 0 0 1 1 0 0 2

Nicrophorus defodiens

Mannerheim 1846 Silph. D 2 11 13 0 1 24 1 26

Nicrophorus hebes Kirby 1837 Silph. D 233 509 226 15 11 932 10 968

Nicrophorus hybridus Hatch & Angell 1925

Silph. D 4 0 1 0 0 5 0 5

137

Nicrophorus marginatus

Fabricius 1801 Silph. D 2 2 18 0 0 22 0 22

Nicrophorus obscurus Kirby 1837 Silph. D 0 3 2 0 0 5 0 5

Nicrophorus orbicollis Say 1825 Silph. D 127 178 177 1 0 478 3 482

Nicrophorus sayi Laporte 1840 Silph. D 0 1 3 0 0 4 0 4

Nicrophorus tomentosus

Weber 1801 Silph. D 30 60 97 0 0 186 1 187

Nitidotachinus tachyporoides

(Horn 1877) Staph. O 1 6 8 3 4 5 3 15

Oiceoptoma noveboracense

(Forster 1771) Silph. D 151 162 90 1 0 402 0 403

Omalium rivulare (Paykull 1789) Staph. D 1 0 0 0 0 1 0 1

Ontholestes cingulatus

(Gravenhorst 1802) Staph. P 0 6 2 0 3 5 0 8

Onthophagus hecate (Panzer 1794) Scarab. D 33 48 76 4 0 150 3 157

Onthophagus nuchicornis

(Linnaeus 1758) Scarab. D 11 7 22 1 1 37 1 40

Pachystilicus hanhami

(Wickham 1898) Staph. P 0 1 1 0 2 0 0 2

Paederus littorarius Gravenhorst 1806 Staph. P 35 31 45 30 30 34 17 111

Paracercyon

minisculus

(Melsheimer 1844) Hydro. D 0 0 1 0 0 1 0 1

Philonthus blossius Smetana 1995 Staph. P 9 8 4 2 6 10 3 21

Philonthus busiris Smetana 1995 Staph. P 3 9 2 2 1 10 1 14

Philonthis caeruleipennis

(Mannerheim 1830)

Staph. P 0 0 1 0 1 0 0 1

Philonthus carbonarius

(Gravenhorst 1802) Staph. P 1 0 0 0 0 1 0 1

Philonthus cognatus (Stephens 1832) Staph. P 5 3 7 2 9 0 4 15

Philonthus couleensis Hatch 1957 Staph. P 0 0 1 0 1 0 0 1

Philonthus cruentatus (Gmelin 1790) Staph. P 1 1 1 1 0 1 1 3

Philonthus lomatus Species Group

Erichson 1840 Staph. P 17 17 37 19 11 26 15 71

Philonthus

occidentalis

Horn 1884 Staph. P 0 2 0 0 0 2 0 2

Philonthus

opacipennis

Notman 1919 Staph. P 1 2 0 1 0 2 0 3

Philonthus palliatus (Gravenhorst 1806) Staph. P 0 1 1 0 0 2 0 2

Philonthus politus (Linnaeus 1758) Staph. P 2 1 1 0 0 4 0 4

Philonthus schwarzi Horn 1884 Staph. P 51 21 31 28 15 45 15 103

Philonthus sericinus Horn 1884 Staph. P 1 0 0 0 0 1 0 1

Philonthus validus Casey 1915 Staph. P 1 2 4 0 0 7 0 7

Philonthus varians (Paykull 1789) Staph. P 17 12 10 0 0 38 1 39

Phyllophaga sp. Harris 1827 Scarab. H 1 0 0 1 0 0 0 1

Pselaphus fustifer Casey 1894 Staph. P 0 0 1 0 0 0 1 1

Pycnoglypta

campbelli

Gusarov 1995 Staph. N/A 0 0 1 0 1 0 0 1

Rhyssemus sonatus LeConte 1881 Scarab. D 0 1 0 1 0 0 0 1

Rugilus biarmatus (LeConte 1880) Staph. P 1 0 1 1 0 1 0 2

Saprinus distinguendis

Marseul 1855 Hister. P 1 0 4 0 0 5 0 5

Saprinus oregonensis LeConte 1844 Hister. P 0 2 0 0 0 2 0 2

Sciodrepoides terminans

(LeConte 1850) Leiod. D 10 24 40 2 0 70 2 74

Sciodrepoides watsoni

(Spence 1815) Leiod. D 13 5 18 0 0 36 0 36

Sphaeridium scarabaeoides

(Linnaeus 1758) Hydro. P 1 0 0 0 0 0 1 1

Staphylinus ornaticauda

LeConte 1863 Staph. P 0 4 7 0 5 5 1 11

Stenus angustus Casey 1884 Staph. P 1 0 0 0 0 0 1 1

Stenus difficilis Casey 1884 Staph. P 1 0 0 1 0 0 0 1

138

Stenus melanarius Stephens 1833 Staph. P 0 2 3 2 0 2 1 5

Stenus morio Gravenhorst 1806 Staph. P 1 0 0 0 0 1 0 1

Stenus undescribed species

Latreille 1796 Staph. P 0 4 3 2 0 3 2 7

Stenus scabiosus Casey 1884 Staph. P 0 0 2 2 0 0 0 2

Stenus schwarzi Casey 1884 Staph. P 0 0 1 0 1 0 0 1

Stictolinus flavipes (LeConte 1863) Staph. P 0 0 2 0 0 2 0 2

Tachyporus atriceps Stephens 1832 Staph. O 0 0 1 0 0 0 1 1

Tachyporus borealis Campbell 1979 Staph. O 0 0 2 0 0 1 1 2

Tachyporus canadensis

Campbell 1979 Staph. O 6 0 16 4 5 3 10 22

Tachyporus nitidulus (Fabricius 1781) Staph. O 0 1 3 2 1 0 1 4

Tachyporus pulchrus Blatcheley 1910 Staph. O 0 0 1 0 0 1 0 1

Tachyporus

rulomoides

Campbell 1979 Staph. O 1 0 0 1 0 0 0 1

Tachyporus

transversalis

Gravenhorst 1806 Staph. O 2 8 4 4 3 5 2 14

Tertartopeus niger (LeConte 1863) Staph. P 0 0 1 0 0 0 1 1

Thanatophilus lapponicus

(Herbst 1793) Silph. D 63 98 67 0 0 228 0 228

Tympanophorus

puncticollis

(Erichson 1840) Staph. P 1 1 1 1 0 2 0 3

Total = 117 species Total

3453 8864 6278 1652 1799 12890 2254 18595

Avg. Total per site

690.6 1477 896.86

Number of Species

73 73 91

Avg. Number of Species per site

14.6 12.17 13

Table 4.13 List of study sites with corresponding disturbance type and coordinates.

Site Disturbance Type Latitude (° N) Longitude (° W) Stefanchuk 2 Undisturbed 49.134385 -96.740645

Smook SW Undisturbed 49.1677425 -96.74031

Smook NW Undisturbed 49.17568 -96.740368

Townsend Undisturbed 49.163795 -96.67451

Forrester & Casson Undisturbed 49.1185725 -96.66499

Dykun Undisturbed 49.1049425 -96.649898

Clough Undisturbed 49.1576575 -96.649858

Stefanchuk Grazed 49.1553 -96.739845

Nickel Grazed 49.1629125 -96.716645

Fisher Grazed 49.149765 -96.695408

Fisher 2 Grazed 49.1416775 -96.695463

Penner Grazed 49.1205975 -96.654183

Storoschuk Grazed 49.13253 -96.650598

Singh NW Burned 49.148185 -96.74066

Saranchuk Burned 49.179495 -96.67225

Rose Burned 49.171915 -96.672625

Kai Kwong Lee Burned 49.17512 -96.650273

Bott Burned 49.16934 -96.65027

139

Chapter 5: Conflicting response of pollinator, decomposer, and predator insects to

disturbance and landscape in northern tallgrass prairie

5.0 Abstract

Tallgrass prairie (TGP) management has typically focused solely on the maintenance and

proliferation of the floral community. There is now increased realization that arthropods in TGP

perform several beneficial tasks that increase the productivity and natural beauty of the prairie

landscape. It is not entirely clear if burning and grazing, two common management practices,

positively or negatively impact prairie arthropods, and whether these benefits or negative

responses are a result of direct effects or indirect effects via changes in floral/soil conditions. I

set out to test how pollinators, represented by ground and stem nesting wild bees; decomposers,

and predators, both represented by the haplogastran beetle community; respond to burning and

grazing in the TGP. As well, I measured landscape characteristics to determine how diversity,

composition, and edge amount in the environment affect these insect guilds. I found that

communities of both bees and beetles were distinct between burned and grazed sites. Overall,

bee abundance, richness, and diversity were all significantly higher in undisturbed sites

compared to grazed sites. Rarefied diversity measures revealed that burned sites were often more

diverse than grazed sites, except for stem nesting bees, which showed lowest diversity in burned

sites. For the entire beetle community, grazed sites supported greater abundance compared to

burned and undisturbed sites, but significantly fewer effective species than burned sites. When

separated by guild, predatory beetles were more abundant in burned sites, while decomposer and

unidentified aleocharine beetle abundance was significantly higher in grazed sites. While bees

responded positively to increased landscape diversity, beetles were unaffected. In both bees and

beetles, abundance, richness, and diversity were all positively influenced by the amount of forest,

ditch, and pasture in the surroundings, or negatively influenced by the amount of prairie and

savannah in the surroundings, though they responded at different scales. The ground cover of

sites influenced bee abundance, and ground-nesting bee abundance and richness, in all cases

responding positively towards increasing forb cover, number of flowers, and increasing bare

ground. These qualities are key elements to bee foraging and nesting. Conversely, beetle richness

and predatory beetle abundance, richness, and diversity were all influenced in the opposite

direction than the bees, towards increasing grass cover, which is consistent with life histories that

aren’t dependent on flowers and underground nests. Considering that most study sites were

140

without management for several years prior to my study, my results seem to imply that

management decisions can have long lasting effects on bee and beetle communities beyond

active management years. The contrasting response of these guilds to disturbance type indicate

that it may be best to maintain a diversity of techniques but ensuring that there are fire and

grazing free refugia for especially vulnerable species. The discovery of new provincial records

and even one new species indicates how understudied the insect fauna of Manitoba remains, and

highlights the importance of maintaining large swaths of the ancestral tallgrass prairie.

5.1 Introduction

Insects perform a variety of beneficial services in tallgrass prairie (TGP) (Whiles and

Charlton 2006). These services include pollination, which influences the floral community

composition; predation, which influences the faunal community composition, and

decomposition, which influences nutrient cycling and how energy is converted in the prairie

(Whiles and Charlton 2006). TGP is commonly managed by several methods, including grazing

and prescribed fire (Sveinson 2003). To test how these disturbance types influence the

abundance and diversity of beneficial prairie insects, I conducted a two-part experiment where I

caught and identified wild bees and haplogastran beetles in sites that differed by their disturbance

history, as well as by their landscape diversity, landcover composition, and ground cover

characteristics. Here I summarize the main results of my findings.

5.2 Disturbance Type Effects

While disturbance type was an important variable in many of the bee models, it mostly

affected beetles via abundance, though for the combined beetle matrix, it also influenced

diversity. In the case of both bees and beetles, community composition was significantly

different between burned and grazed sites. For bees, undisturbed sites showed increased

abundance, richness, and diversity compared to grazed sites (Figure 5.1). When bees were

separated by nesting guild, undisturbed sites supported greater ground nesting species richness

than grazed sites, while burned sites supported greater abundance of stem nesting bees than

grazed sites. Rarefied richness curves demonstrated that grazed sites were overall worse for bee

diversity in most groups, while burned sites showed reduced diversity of stem nesters compared

to undisturbed sites according to the rarefied Hill’s Shannon and Inverse Simpson’s graphs. For

141

beetles, grazed sites showed increased abundance of beetles compared to burned and undisturbed

sites for both the total haplogastran beetles (Figure 5.2) and the aleocharines, however, burned

sites housed significantly higher diversity of beetles compared to grazed and undisturbed sites

(Figure 5.3). When identified beetles were separated by feeding guild, burned sites had a greater

abundance of predatory beetles than either grazed or undisturbed sites, while grazed sites showed

higher abundance of decomposer beetles than undisturbed sites. The rarefied diversity curves

confirmed that burned sites were more diverse than grazed and undisturbed sites in all but the

predatory guild. It has been pointed out that the positive effects of grazing probably only extend

to those insect guilds which directly benefit from it through increasing dung input (van Klink et

al. 2015). Therefore, while decomposer beetles may have benefitted in abundance through cattle

grazing or through an increased organic matter litter layer in my study, it is important to not

extrapolate this positive effect to other trophic levels.

Figure 5.1 Bar graph representing the effect of the three disturbance types on bee bowl caught

bee abundance, species richness, and Hill’s Shannon effective species number. Error bars

represent the standard error. Different letters above bars represent significant differences based

on GLMs with disturbance type as the explanatory variable with a negative binomial

(abundance), quasi-Poisson (richness), and Gamma (Hill’s Shannon) distributions (alpha = 0.05).

a

a a

a a

b a b

ab

142

Figure 5.2 Bar graphs representing the effect of the three disturbance types on beetle abundance

in 2017 (left) and 2018 (right). Error bars represent the standard error. Different letters above the

bars represent significant differences based on GLMs with disturbance type as the explanatory

variable with negative binomial distributions (alpha=0.05).

Figure 5.3 Bar graphs representing the effect of the three disturbance types on beetle species

richness in 2017 and 2018, and on Hill’s Shannon effective species numbers in 2017 and 2018.

a

b b

a a

a ab b

a

a a a

b

a a

a

b

c

143

Error bars represent the standard error. Different letters above the bars represent significant

differences based on GLMS with disturbance type as the explanatory variable and Poisson

(richness) and Gamma (Hill’s Shannon) distributions (alpha=0.05).

5.3 Landscape Composition Effects

While bees responded most at scales of 1 km, which is at the high end of the typical

foraging range for wild bees (Gathmann and Tscharntke 2002, Zurbuchen et al. 2010), beetles

responded more at the 500 m scale, which might reflect some of the flightless or rare to fly

species. Bees universally responded positively to increased landscape diversity, while beetles did

not respond at all. Bees and beetles responded similarly to landcover composition, with both

having positive effects of the landcover PCA, indicating that they are either responding

negatively to increasing prairie and savannah in the environment, or positively to increasing

forest, ditch, and pasture cover. The northern block of the MTGPP is very wet, which could

explain a negative effect of increasing prairie cover. As well, bees are known to respond

positively to increasing forest cover in TGP (Griffin et al. 2017, 2021). Predatory beetles are also

known to be more diverse in forest adjacent to TGP (Roughley et al. 2006), though I found

positive effects mostly in my decomposer beetle models, which conflicts with another prairie

study that found decreased silphid abundance under increasing forest cover in a prairie

ecosystem (Walker Jr and Hoback 2007). There is some evidence that increased forest area and

edge benefits abundance and richness of rove beetles in grassland ecosystems (Tóthmérész et al.

2014), which could potentially explain my results. Decomposer beetle abundance was positively

affected by increased amount of edge in the environment. This conflicts with some other findings

that found negative (Rocha-Ortega and Coronel-Arellano 2019) or negligible (Sánchez-de-Jesús

et al. 2016) effects of patch shape, but they were conducted mostly in subtropical environments

and looking specifically at dung beetles.

5.4 Ground Cover Effects

Bees and beetles both responded to ground cover effects, but in opposite ways. When

present in bee models, the ground cover PCA had a positive effect, implying that increased forb

144

cover, bare ground cover, number of flowers, and decreased amount of grass cover benefitted

bee abundance and diversity, especially for the ground nesting guild. Predatory beetles on the

other hand, showed a negative effect of the PCA in their models, implying that increased grass

cover positively affected their diversity and abundance. This is consistent with another grassland

study that found the same effect (Crist and Peters 2014).

5.5 New and Interesting Records

We caught at least two bee provincial records, Melissodes desponsus Smith, and

Lasioglossum (Dialictus) michiganense (Mitchell). The former is a pollen specialist on thistle,

Cirsium spp., while the latter is a social parasite of other Dialictus bees (Gibbs 2011). For

beetles, I caught 19 new records: two hydrophilids, three leiodids, and 14 staphylinids. I caught

one beetle that Adam Brunke (CNC) believes has not been previously noted or described from

the genus Stenus. Adam Brunke (CNC) also discovered a previously undescribed species of

Ischnosoma based on my collections, though it has previously been caught elsewhere and just

misidentified. A few of the records are of very minute rove beetles, including Micropeplus

sculptus LeConte, and Euconnus flavitarsis (LeConte). Four of the new records belong to the

large, predatory rove beetle genus Philonthus, which are largely unidentifiable without dissection

of male genitalia, so have perhaps previously been overlooked. I also caught 13 specimens of

Staphylinus ornaticauda LeConte, a species that has been suggested as being vulnerable to

extinction given that it is large and flightless and has a relatively narrow habitat range (Brunke et

al. 2011). It is also listed as critically imperiled in the province of Ontario according to

NatureServe, but it is unranked in Manitoba.

5.6 Future Directions

Since historically TGP experienced combinations of fire and grazing, a study in the

southern block of the MTGPP (where both practices are used in combination on individual sites)

should be undertaken to determine how bees and beetles respond when both practices are in use.

Future studies should also measure stocking rate and mowing/haying regimes of sites, to see if

they are confounding the results. As well, the preserve should attempt to allow bison to graze on

some sites, since they would have historically been present, and they are known to have different

effects on prairie floral (Towne et al. 2005) and faunal elements (Rosenberger and Conforti

145

2020) than cattle. Where possible, other beneficial insect guilds should be studied on the prairie,

especially so in the case of syrphid flies, who act as both pollinators and predators (and

sometimes even as saprophages and pollinators) and are known to be the most important

pollination group in northern TGP (Robson 2008). Many studies mentioned the importance of

measuring soil characteristics to better understand grassland insect fauna, so future studies in

TGP should always collect as many soil characteristics as possible, as they may be playing a

large role in insect abundance and diversity.

5.7 Conclusion

It is imperative that habitat and ecosystem restoration programs move beyond a strictly

plant focused framework (Lengyel et al. 2020). Future conservation should focus on increasing

habitat diversity, via a range of different disturbances/lack of disturbances, to maximize TGP

animal diversity (Lengyel et al. 2020). This study shows that different beneficial insect taxa, and

discrete guilds therein, respond in diverse ways to disturbance, landscape, and ground cover.

While current management strategies may benefit one group, they may be inadvertently harming

another. It is hoped that this study can influence the MTGPP management committee, along with

other TGP preserves, to increase efforts to study the effects of management and landscape

context on insect fauna. Insects are too often neglected in restoration programs, despite the wide

variety of functions that they perform in the TGP (Whiles and Charlton 2006).

5.8 References

Brunke, A., Newton, A., Klimaszewski, J., Majka, C., and Marshall, S. 2011. Staphylinidae of eastern Canada and adjacent United States. Key to subfamilies; Staphylininae: tribes and

subtribes, and species of Staphylinina. Canadian Journal of Arthropod Identification,: 1–110. doi: 10.3752/cjai.2011.12.

Crist, T.O., and Peters, V.E. 2014. Landscape and local controls of insect biodiversity in conservation grasslands: implications for the conservation of ecosystem service providers in

agricultural environments. Land, 3: 693–718. doi: 10.3390/land3030693.

Gathmann, A., and Tscharntke, T. 2002. Foraging ranges of solitary bees. Journal of Animal Ecology, 71: 757–764.

Gibbs, J. 2011. Revision of the metallic Lasioglossum (Dialictus) of eastern North America

(Hymenoptera: Halictidae: Halictini). In Zootaxa. doi: 10.11646/%x.

Griffin, S.R., Bruninga-Socolar, B., and Gibbs, J. 2021. Bee communities in restored prairies are structured by landscape and management, not local floral resources. Basic and Applied Ecology, In Press: 1–8.

146

Griffin, S.R., Bruninga-Socolar, B., Kerr, M.A., Gibbs, J., and Winfree, R. 2017. Wild bee community change over a 26-year chronosequence of restored tallgrass prairie. Restoration Ecology, 25: 650–660. doi: 10.1111/rec.12481.

van Klink, R., van der Plas, F., (Toos) van Noordwijk, C.G.E., WallisDeVries, M.F., and Olff, H. 2015. Effects of large herbivores on grassland arthropod diversity. Biological Reviews, 90: 347–366. doi: 10.1111/brv.12113.

Lengyel, S., Mester, B., Szabolcs, M., Szepesváry, C., Szabó, G., Polyák, L., Boros, Z., Mizsei,

E., Málnás, K., Méro, T.O., and Aradi, C. 2020. Restoration for variability: emergence of the habitat diversity paradigm in terrestrial ecosystem restoration. Restoration Ecology, 28: 1087–1099. doi: 10.1111/rec.13218.

Robson, D.B. 2008. The structure of the flower–insect visitor system in tall-grass prairie.

Botany, 86: 1266–1278. doi: 10.1139/B08-083.

Rocha-Ortega, M., and Coronel-Arellano, H. 2019. How predictable are the responses of ant and dung beetle assemblages to patch and landscape attributes in fragmented tropical forest landscapes? Landscape and Ecological Engineering, 15: 315–322. doi: 10.1007/s11355-

018-0367-9.

Rosenberger, D.W., and Conforti, M.L. 2020. Native and agricultural grassland use by stable and declining bumble bees in Midwestern North America. Insect Conservation and Diversity,. doi: 10.1111/icad.12448.

Roughley, R.E., Pollock, D.A., and Wade, D.J. 2006. Biodiversity of ground beetles (Coleoptera: Carabidae) and spiders (Araneae) across a tallgrass prairie – aspen forest ecotone in southern Manitoba. The Canadian Entomologist, 138: 545–567.

Sánchez-de-Jesús, H.A., Arroyo-Rodríguez, V., Andresen, E., and Escobar, F. 2016. Forest loss

and matrix composition are the major drivers shaping dung beetle assemblages in a fragmented rainforest. Landscape Ecology, 31: 843–854. doi: 10.1007/s10980-015-0293-2.

Sveinson, J.M.A. 2003. Restoring tallgrass prairie in southern Manitoba, Canada. [degree] thesis, University of Manitoba, .

Tóthmérész, B., Nagy, D.D., Mizser, S., Bogyó, D., and Magura, T. 2014. Edge effects on ground-dwelling beetles (Carabidae and Staphylinidae) in oak forest-forest edge-grassland habitats in Hungary. European Journal of Entomology, 111: 686–691. doi: 10.14411/eje.2014.091.

Towne, E.G., Hartnett, D.C., and Cochran, R.C. 2005. Vegetation trends in tallgrass prairie from bison and cattle grazing. Ecological Applications, 15: 1550–1559.

Walker Jr, T.L., and Hoback, W.W. 2007. Effects of invasive Eastern Redcedar on capture rates of Nicrophorus americanus and other Silphidae. Environmental Entomology, 36: 297–307.

Whiles, M.R., and Charlton, R.E. 2006. The ecological significance of tallgrass prairie arthropods. Annual Review of Entomology, 51: 387–412. doi: 10.1146/annurev.ento.51.110104.151136.

Zurbuchen, A., Landert, L., Klaiber, J., Müller, A., Hein, S., and Dorn, S. 2010. Maximum

147

foraging ranges in solitary bees: only few individuals have the capability to cover long foraging distances. Biological Conservation, 143: 669–676. doi: 10.1016/j.biocon.2009.12.003.