INFLUENCE OF LIVING AND NON-LIVING HABITAT COMPLEXITY ON
ARTHROPODS IN STRIP-TILLED CABBAGE FIELDS
By
Alexandria Bryant

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Entomology - Master of Science
2013

ABSTRACT
INFLUENCE OF LIVING AND NON-LIVING HABITAT COMPLEXITY ON
ARTHROPODS IN STRIP-TILLED CABBAGE FIELDS
By
Alexandria Bryant
Vegetational (living) and structural (non-living) complexity can influence arthropods in
agricultural systems. In this thesis, both forms of complexity were manipulated in cabbage fields.
In 2010 and 2011, a field experiment was conducted examining the effects of herbicide intensity
and oat kill date on arthropods, weeds, and crop yield. Reduced herbicide intensity was
associated with 1) lower abundance of several key cabbage pests and, 2) greater abundance of
important natural enemy species. Habitats with both later oat kill dates or reduced herbicide
intensity contained 1) fewer herbivores with chewing feeding guilds and more specialized diet
breadths and, 2) greater abundance of active hunting natural enemies. In 2012, mulch species and
mulch management practices were examined. Although mulching did not provide significant pest
suppression, several generalist and one specialist parasitoid responded positively to habitat
complexity. Predation and parasitism rates determined from natural enemy exclusion cages
indicated no impact of mulching on the biocontrol of a major cabbage pest, the imported
cabbageworm. In 2012, natural enemy behavior was observed in a y-tube olfactometer to
determine the role of odor cues from decaying mulch on habitat choice. In general, natural
enemies were not attracted or repelled by mulch odors. Further work is needed before
implementation of these habitat management strategies (e.g. mulching and reduced weed
management) and caution should be taken to avoid reducing cabbage yields or enhancing
hyperparasitism.

ACKNOWLEDGEMENTS

First, I would like to thank my advisor, Zsofia Szendrei, for her guidance and
support at each stage of my graduate work and the experiences in outreach and extension that she
provided. Those experiences have led to the career path I have chosen. I also greatly appreciate
the guidance and helpful feedback I received from my graduate committee members, Dan
Brainard, Jen Lau, and Matt Grieshop over the past three years. I appreciate their patience,
understanding, and expertise. To my lab mates, Rob Morrison, Diana Londono, and Adam Byrne,
thanks for making my graduate school experience unforgettable. I could never have made it this
far without their support, humor, and friendship. Thanks for always being a shoulder to lean on, a
brain to pick, and an ear to listen to my chatter. Thanks for keeping me sane.
The work in this thesis would not have been possible without the efforts of a large group
of people. Thanks, to Bill Chase, Corey Noyes, and to all the Hort Farm employees who helped
till, spray, maintain, and irrigate cabbage plots. To all the undergraduate students from the
Szendrei and Brainard labs who aided in data collection and helped weed cabbage plots, I greatly
appreciate your hardwork. I’m also thankful for the hardwork and patience of the Entomology
office staff members. Also, thanks to my funding sources, the Plant Science Fellowship at MSU
and Project GREEEN which made these projects possible.
Finally, I would like to thank my family for their love and support. I dedicate this thesis
to them. In particular, I dedicate this work to my little neice and nephew who always remind me
to keep learning, keep asking questions, and never lose the excitement that comes with even the
smallest discovery. I also dedicate this thesis to my father, who has been an endless source of
strength, love, and support.

iii

TABLE OF CONTENTS

List of Tables……………………………………………………………………………….…....vii
List of Figures………………………………………………………………………………..…...ix
CHAPTER 1: Habitat complexity and managing arthropods in cabbage……………….………...1
Introduction..……..……………………………………………………………………......1
Insect pests in cabbage…….....……………………………....…………………………....2
Chemical control in cabbage fields………………………………….…………………….4
Natural biological control in cabbage fields………………………………………………5
Habitat complexity and conservation biological control………………………...………..8
Vegetational and structural complexity in Michigan cabbage……...………………..........8
Mechanisms for response of natural enemies to habitat complexity….…….……..……..10
Thesis objectives………………………………………………….……………………....14
CHAPTER 2: Cover crop mulch and weed management influence arthropod communities in
strip-tilled cabbage……………………………………………………………………………….15
Introduction……………………………………………………………...……………….15
Materials and Methods…………………………………………………………………...21
Experimental Design………………………………………………...…………...21
Field Operations……………………………………………………...…………..21
Oat height and biomass…………………………………………………………..23
Weed biomass……………………………………………………………..……..24
Cabbage biomass and yield……………………………………………………....24
Habitat and Cabbage Yield Statistical Analysis………………………………....25
Insect community assessment……………………………………………………25
Results…………………………………………………………………………………....27
Oat height and biomass………………………………………………………......27
Weed biomass…………………………………………………………………....28
Cabbage biomass and yield………………………………………………………28
Insect community structure………………………………………………………30
Insect abundance…………………………………………………………………31
Impact of oat kill date on insect abundance…………………....…….......31
Impact of herbicide intensity on insect abundance…………....…………32
Significant covariates for herbivores…...…..…………………………….33
Significant covariates for natural enemies ....……………………………33
Insect functional groups…………………………………………………….........34
Discussion………………………………………………………………………………..45
Insect community structure…………………………………………………........45
Insect abundance…………………………………………………………………45
Impact of oat kill date on insect abundance………………...………........46
Impact of herbicide intensity on insect abundance…………....……........47
Significant covariates for herbivores………….………………………….48

iv

Significant covariates for natural enemies…………………………….….49
Insect functional groups………………………...………………………..………50
Summary…………….…………………………………………………....……...51
CHAPTER 3: The impact of mulch species and management on natural enemies and biological
control in cabbage……………….……………………………………………………………….53
Introduction……………………………………………………………………………....53
Materials and Methods………….………………………………………………………..56
Experimental Design……………………………………………………………..56
Field Operations………………………………………………………………….57
Mulch, weed, and cabbage yield measurements………………..………………..58
Mulch, weed, and cabbage yield statistical analysis………………..……………59
Arthropod sampling………………………………………………………...……59
Arthropod communities………………………………………………………….60
Arthropod abundance……………………………………………………………61
Natural enemy exclusion cages……………………………………………….…61
Results……………………………………………………………………………………63
Mulch and cabbage yield measurements…………………………………..…….63
Arthropod communities……………………………………………………….…65
Arthropod abundance……………………………………………………………66
Significant covariates for herbivores……………………...……………………..67
Significant covariates for natural enemies………………...……………………..68
Natural enemy exclusion cages………………………………………………..…68
Mortality.………………………………………..……………………..…68
Parasitism………………………………………..…………………….…69
Discussion……………………………………………………….……………………….86
Mulch and cabbage yield measurements………………………………………...86
Arthropod communities……………………………………...………………......86
Arthropod abundance…………………………………………………………....87
Significant covariates for herbivores…………………………………………….90
Significant covariates for natural enemies……………………………………….90
Natural enemy exclusion cages…………………………………...……………...91
Summary……………………….………………………………………………………...92
CHAPTER 4: The role of olfactory cues from cover crop mulch on two generalist and one
specialist natural enemy in cabbage……………………………………………………………...94
Introduction…………………………………………………………..………………..…94
Materials and Methods………………………………………………...…………………97
Y-tube olfactometer……………………………………………..…………….…97
Insects ………………………………………………………………...…97
Odor sources…………………………………………….…………….....98
Y-tube assay…………………………………………….………………100
Results………………………………………………………………….……………….101
Y-tube olfactometer……………………………………………..……………101
Conclusions……………………………………………………………….…………….106
Y-tube olfactometer………………………………………………..…………106

v

Summary…………………………………………………………….………….107
CHAPTER 5: Conclusions and Future Research……………………...………………….….…108
APPENDICES………………………………………………………………………………….111
Appendix 1.1: Supplementary data……………………………………………………112
Appendix 1.2: Record of deposition of voucher specimens………………………..….121
LITERATURE CITED…………………………………………………………………………123

vi

LIST OF TABLES
Table 2.1. Dates of major field operations in 2010 and 2011………………………………........23
Table 2.2. Mean ± SEM oat height, oat biomass, and weed biomass in 2010 and 2011………...36
Table 2.3. Mean ± SEM non-marketable and marketable fresh weight of cabbage heads in 2010
and 2011………………………………………………………………………………….............37
Table 2.4. Mean ± SEM herbivore abundance in experimental plots across fixed effect: oat kill
date (early, intermediate, or late) in 2010 and 2011……………………………………..............38
Table 2.5. Mean ± SEM herbivore abundance in experimental plots across fixed effect: herbicide
intensity (low or high) in 2010 and 2011………………………………………….......................39
Table 2.6. Mean ± SEM natural enemy abundance in experimental plots across fixed effect: oat
kill date (early, intermediate, or late) in 2010 and 2011……………………………...………… 40
Table 2.7. Mean ± SEM natural enemy abundance in experimental plots across fixed effect:
herbicide intensity (low or high) in 2010 and 2011…………………………………………...... 42
Table 3.1. Mean ± SEM mulch height at an early, intermediate, and late date during the growing
season……………………………………………………………………………….....................71
Table 3.2. Mean ± SEM mulch biomass at an early, intermediate, and late date during the
growing season………………………………………………………….…………......................72
Table 3.3. Mean ± SEM weed height, weed abundance, and weed species richness across mulch
species and management in 2012.…………………………………………………......................73
Table 3.4. Mean SEM non-marketable and marketable fresh weight of cabbage heads at
harvest……………………………………………………….…………………………………...74
Table 3.5. Herbivore abundance in experimental plots across fixed effect: mulch management
(no crimping or crimped)………………………………………………………………………...75
Table 3.6. Natural enemy abundance in experimental plots across fixed effect: mulch
management (no crimping or crimped)………………………………………………………….76
Table 3.7. Herbivore abundance in experimental plots across fixed effect: mulch species (bare
soil, hairy vetch, rye)…………………………………………………………………………….78
Table 3.8. Natural enemy abundance in experimental plots across fixed effect: mulch species
(bare soil, hairy vetch, rye)………………………………………………………………………79

vii

Table 4.1. List of treatments or choice tests performed for each natural enemy species (spined
soldier bug, convergent lady beetle, and Cotesia rubecula)……………………………………..99
Supplementary Table S1. Arthropods were categorized according to feeding functional group,
feeding guild, diet breadth, and hunting mode in 2010 and 2011……………...……………….112
Supplementary Table S2. Mean ± SEM cabbage biomass, leaf area, and number of leaves per
plant mid-season in 2010 and 2011…………………………………………………..………...114
Supplementary Table S3. Mean ± SEM number of marketable and non-marketable cabbage
heads per plot in 2010 and 2011……………………………………………………………….115
Supplementary Table S4. Mean ± SEM number of non-marketable and marketable cabbage
heads at harvest in 2012………………………………………………………………………..116
Supplementary Table S5. Voucher specimens deposited at the Albert J. Cook Arthropod
Research Collection (Michigan State University)……………………………………………..121

viii

LIST OF FIGURES
Figure 1.1. Photo: Major pests in Michigan cabbage………………………………………….….3
Figure 1.2. Photo: Common natural enemies in Michigan cabbage……………………....………7
Figure 1.3. Direct impacts of habitat complexity on trophic levels in cabbage fields…………...12
Figure 1.4. Indirect impacts of habitat complexity on trophic levels in cabbage fields…………13
Figure 2.1. Habitat complexity dendograms based on average linkage distance for the herbivore
community (A) and natural enemy community (B) in 2010 and 2011……………………….….43
Figure 2.2. Effect of habitat complexity on insect abundance classified according to functional
feeding groups, feeding guilds, diet breadth, and hunting modes……………………………….44
Figure 3.1. Total predator abundance on yellow sticky traps throughout the cabbage growing
season in experimental plots with different mulch species (bare soil, rye, vetch) and mulch
management treatments (no crimping, crimping)……………………….……………………….81
Figure 3.2. Total ground dwelling predators throughout the cabbage growing season in
experimental plots with different mulch species (bare soil, rye, vetch) and mulch management
treatments (no crimping, crimping)………………………………………………………...……82
Figure 3.3. Total parasitoid abundance on yellow sticky traps throughout the cabbage growing
season in experimental plots with different mulch species (bare soil, rye, vetch) and mulch
management treatments (no crimping, crimping)…………………….………………………….83
Figure 3.4. Percent predation (Mean ± SEM) of the imported cabbageworm across natural enemy
exclusion cages…………………………………………………………………………………..84
Figure 3.5. Percent parasitism (Mean ± SEM) of the imported cabbageworm across natural nemy
exclusion cages………...…………………………………………………….…………………..85
Figure 4.1. Odor sources for y-tube olfactometer………………………….…………………...100
Figure 4.2. Response of spined soldier bug nymphs in Y-tube olfactometer choice test……....103
Figure 4.3. Response of convergent lady beetles in Y-tube olfactometer choice test………….104
Figure 4.4. Response of Cotesia rubecula wasps in Y-tube olfactometer choice test...…….….105
Figure S1. Community dendogram of plots based on average linkage distance for herbivores in
2010 (A) and 2011 (B) and natural enemies in 2010 (C) and 2011 (D)………………………..117

ix

CHAPTER 1.
Habitat complexity and managing arthropods in cabbage

Introduction. Plants in the family Brassicaceae are considered one of the oldest groups of
cultivated plants in the world, with archaeological evidence of human use dating back to 5000
BC (Ahuja et al. 2011). Even today, these plants are playing an important role in global
agriculture. In 2010, global production of Brassica sp. exceeded 2.25 million hectares and
yielded over 66.4 million tons (FAO, 2010). This family contains a variety of plants considered
crops, ornamentals, and weeds with uses ranging from food for humans and livestock to oil,
medicines, and renewable bio-energy. Among the Brassica crops commonly harvested, global
production of cabbage (Brassica oleracea) is greater than any other Brassica crop grown for
human consumption (Ahuja et al. 2011).
The United States has consistently been ranked in the top six countries in cabbage
production (USDA-NASS, 2011). In 2011, 66,900 acres of cabbage were planted in the United
States for the fresh market. Total production of cabbage in 2011 was greater than 21 million cwt,
valued at $368,311,000 (USDA-NASS, 2011).
Michigan has a long history of vegetable production, with a great diversity of fresh
th

market vegetables, including Brassica crops. In the United States, Michigan is ranked 10 in
fresh market cabbage production with 759 thousand cwt of fresh market cabbage produced in
2011 (NASS). With the majority of Michigan cabbage being grown for fresh market, threats
from diseases, weeds, and insects pose severe challenges to achieving marketable cabbage heads.
Weather conditions and the loam soils in southern Michigan are ideal for cabbage production.

1

Insect pests in cabbage. Marketability of cabbage heads declines when insect feeding reduces
biomass, contaminates the harvestable parts with frass, and causes cosmetic damage. The
diversity of insect pest species attacking Michigan cabbages poses a challenge to maintain
quality with economical and sustainable tools. Insects such as the diamondback moth (Plutella
xylostella, Lepidoptera: Plutellidae), imported cabbageworm (Pieris rapae, Lepidoptera:
Pieridae), cabbage looper (Trichoplusia ni, Lepidoptera: Noctuidae), crucifer flea beetle
(Phyllotreta cruciferae, Coleoptera: Chrysomelidae), and striped flea beetle (Phyllotreta
striolata, Coleoptera: Chrysomelidae) chew characteristic holes, damaging cabbage leaves and
heads (Figure 1.1). Other insects, such as cabbage aphids (Brevicoryne brassicae, Hemiptera:
Aphididae), green peach aphids (Myzus persicae, Hemiptera: Aphididae), onion thrips (Thrips
tabaci, Thysanoptera: Thripidae), and white flies (Hemiptera: Aleyrodidae), pierce cabbage
tissue and remove plant sap (Figure 1.1). This feeding can reduce plant growth, cause
characteristic leaf chlorosis and leaf curling, and introduce or facilitate plant diseases such as
molds (Leite et al. 2006). In addition to reducing marketable yield, damage from insect pests can
lead to 90% crop losses in severe cases (Ahuja et al. 2011). Management strategies to prevent
yield and crop losses often focus on individual pests or a complex of pests. In cabbage,
lepidopteran pests (e.g. diamondback moth, imported cabbageworm, and cabbage looper) cause
the greatest damage in Michigan cabbage and management strategies often focus on these pests.
The current global expense for managing one of these lepidopteran pests, the diamondback moth
(Plutella xylostella, Lepidoptera: Plutellidae), and crop losses due to this pest, exceeds a billion
dollars a year (Grzywacz et al 2010).

2

A)

B)

C)

D)

E)

F)

G)

H)

Figure 1.1. Major pests in Michigan cabbage. Top row: diamondback moth (Plutella xylostella,
Lepidoptera: Plutellidae) larva (A) and adult (B), Second row: imported cabbageworm (Pieris
rapae, Lepidoptera: Pieridae) larva (C) and adult (D), Third row: cabbage looper (Trichoplusia
ni, Lepidoptera: Noctuidae) larva (E) and adult (F), Bottom row: striped flea beetle (Phyllotreta
striolata, Coleoptera: Chrysomelidae) (G) and cabbage aphids (Brevicoryne brassicae,
Hemiptera: Aphididae) (H). For interpretation of the references to color in this and all other
figures, the reader is referred to the electronic version of this thesis.

3

Chemical control in cabbage fields. Traditionally, conventional management practices for
insect pests in cabbage include pesticide applications according to calendar dates without
considering thresholds (Hutchison and Burkness 1999). In addition to costs associated with this
overuse of insecticides, current chemical controls have led to development of insecticide
resistance to major insecticides commonly included in cabbage spray schedules (e.g. pyrethroids
and carbamates) (Leibee and Savage 1992). In the United States, insecticide resistance has been
observed in major cabbage pests, such as the cabbage looper and diamondback moth, creating
challenges for growers relying primarily on chemical controls (Wilkinson et al. 1983, Leibee and
Savage 1992). Even Bt sprays that are often considered better alternatives to synthetic
insecticides due to their high specificity and limited environmental impacts, have led to
insecticide resistance. This phenomenon has been observed in diamondback moth and cabbage
looper populations (Talekar and Shelton1993, Janmaat and Myers 2003).
Chemical controls have negative impacts on the ecosystem because agricultural fields are
open systems, connected to the surrounding landscape. Chemical inputs such as fertilizer and
pesticides do not remain in the field. Run off and drift from pesticide applications can cause
mortality of terrestrial and aquatic organisms in adjacent ecosystems (e.g. Burn 2003), as well as
chronic impacts on surrounding wildlife (e.g. Poletta et al. 2009). Negative impacts of pesticide
use can also be observed within the agroecosystem itself. In particular, non-target organisms that
would otherwise provide free pest management and pollination services can be negatively
impacted (Haseeb and Amano 2002, Desneux et al. 2007, Pekar and Benes 2008, Johnson et al.
2010).
Concerns over insecticide resistance and environmental impacts of widespread pesticide
use have led scientists to look for alternative management strategies for cabbage pests and many

4

growers are turning to Integrative Pest Management (IPM) strategies. IPM emphasizes the use of
pest monitoring, appropriate pest thresholds, selective insecticides, and insecticide rotation to
reduce development of resistance. IPM strategies also incorporate other modes of pest
management, including biological control.

Natural biological control in cabbage fields. Although there is a diverse community of insect
pests attacking Michigan cabbage, there is also a diverse community of natural enemies
providing natural biological control in these fields. Common predators include lady beetles
(Coleoptera: Coccinellidae), ground beetles (Coleoptera: Carabidae), predatory thrips
(Thysanoptera: Aeolothripidae), damsel bugs (Hemiptera: Nabidae), big-eyed bugs (Hemiptera:
Geocoridae), spiders (Araneae), and harvestman (Opiliones) (Figure 1.2). Parasitoids also
provide pest suppression, including tachinid flies (Diptera: Tachinidae) and specialists such as
Diadegma insulare (Hymenoptera: Ichneumonidae) and Cotesia rubecula (Hymenoptera:
Braconidae) that attack the diamondback moth and imported cabbageworm respectively (Figure
1.2).
Natural enemies can provide substantial biological control in cabbage fields. Studies have
found significant mortality of the imported cabbageworm in cabbage fields, with predation rates
up to 80% (Schmaedick and Shelton 1999). Long term studies in Minnesota have observed
parasitism rates of this pest as high as 50% (Wold-Burkness et al. 2005). These ecosystem
services are provided by arthropods naturally found in the field. However, insecticide
applications can reduce natural enemy abundance, predation, and parasitism in cabbage fields
(Bommarco et al. 2011), often leading to greater pest populations and investment in pest
management by growers. Secondary pest outbreaks also result from the loss of natural enemies.

5

Although usually considered a minor threat to vegetable crops, without the natural enemies that
usually suppress their populations, secondary pests can cause severe damage and crop loss.

6

A)

E)

J)

B)

C)

G)

F)

K)

H)

L)

D)

I)

M)

Figure 1.2. Common natural enemies in Michigan cabbage. Top row: lady beetles (Coleoptera:
Coccinellidae)(A-D), Second row: predatory thrips (Thysanoptera: Aeolothripidae)(E), damsel
bugs (Hemiptera: Nabidae)(F), big-eyed bugs (Hemiptera: Geocoridae)(G), minute pirate bug
(Hemiptera: Anthocoridae)(H), lacewing larva (Neuroptera: Chrysopidae)(I), Third row: spiders
(Araneae)(J-K), Diadegma insulare (Hymenoptera: Ichneumonidae)(L) and Cotesia rubecula
(Hymenoptera: Braconidae)(M).

7

Habitat complexity and conservation biological control. Natural enemy populations can be
conserved by manipulating habitat. While conventional cabbage management practices may
reduce natural enemy populations, my thesis aimed to examine alternative habitat management
techniques that increase habitat complexity within a field and in turn, enhance biological control
of cabbage pests. Past studies have primarily examined habitat management strategies such as
vegetation borders, intercropping, and undersowing cabbage with living mulches to create
habitat complexity.
Non-crop borders include the planting of vegetational strips along the borders of an
agricultural field or the presence of natural vegetation adjacent to field margins. Much of this
work has examined the impact of floral borders in reducing pest populations and enhancing
natural enemy populations and biological control (Bigger and Chaney 1998, Lee and Heimpel
2005, Winkler et al. 2010), with particular attention to the relationship between floral resources
and effects on parasitoids (Lee et al. 2006, Lee and Heimpel 2008).
Habitat management such as intercropping and undersowing includes the planting of a
cash crop with a companion plant or a cover crop, increasing the complexity within the field. In
cabbage, intercropping can reduce pest populations (Theunissen et al. 1995, Vidal 1997, Lynch
and Zobel 2002, Hooks and Johnson 2003, Xu et al. 2010) and lead to greater natural enemy
populations (Armstrong and McKinlay 1997). In this thesis I also examined alternative
management strategies for creating habitat complexity within the field, such as reduced herbicide
management, strip tillage, and cover crop mulching.

Structural and vegetational complexity in Michigan cabbage. Habitat complexity can be
separated into non-living and living complexity (Rypstra et al. 1999, Finke and Denno 2002).

8

Based on these measurements, two components create habitat complexity. First, structural
complexity includes plant architecture, size, and height of non-living structures such as mulch.
Second, vegetational complexity includes properties of living vegetation, such as biomass and
plant diversity (McNett and Rypstra 2000, Shrewsbury and Raupp 2000, Obermaier et al. 2008).
In this thesis, I examined the effects of both structural complexity (e.g. amount of mulch in
between cabbage rows) and vegetational complexity (e.g. amount of weed biomass in between
cabbage rows) on arthropod communities.
Reducing herbicide intensity increases the weed biomass in between cabbage rows,
therefore creating vegetational complexity in the field. In other Brassica crops, greater
vegetational complexity due to weeds can reduce pest populations, although this is highly
dependent on weed species (Schellhorn and Sork 1997). Combining strip tillage and killed cover
crops creates structural complexity. Strip tillage is a form of reduced tillage in which planting
rows are tilled, leaving soil undisturbed between rows. After killing a fall or spring planted cover
crop, strip tilling of the field leaves cover crop mulch on the soil surface between planting rows.
Benefits of allowing cover crop mulch to remain on the soil surface in cabbage fields have been
observed, including maintaining soil moisture, reducing soil temperatures (Mochizuki et al.
2008), and providing weed suppression (Bottenberg et al. 1997).
Structural complexity can be further manipulated by choosing cover crops with different
physical properties (e.g. leaf area, biomass) or manipulating the physical structure with further
management techniques (e.g. rolling, crimping). Cover crop species such as legumes and cereals
have different physical properties that have influenced cabbage yield, cabbage quality, weed
suppression, and insect abundance (Morse and Seward 1986, Nelson et al. 1991, Mangan et al.
1995, Mwaja et al. 1996, Masiunas et al. 1997). Rolling and chopping these mulches have also

9

had a significant negative and neutral impact on cabbage yield respectively (Mochizuki et al.
2008).
Although the effects of weed biomass, the presence of mulch, and mulch management on
weed suppression and cabbage yields have been examined, limited research has examined the
impacts across multiple trophic levels especially as it pertains to biological control.

Mechanisms for response of natural enemies to habitat complexity. Theoretically, complex
habitats should reduce pest populations directly by acting as a trap crop and interfering with the
movement, landing, and oviposition of pests into the field or attracting the pest away from the
cash crop (Andow 1988, Finch and Collier 2000, Broad et al. 2008). However, complex Brassica
habitats may also provide additional resources for pests such as nectar and pollen, directly
influencing arthropod abundance (Zhao et al. 1992).
According to the natural enemy hypothesis (Elton 1958), complex habitats can provide
refuge, alternative prey, additional resources, and protection from intraguild predation,
enhancing natural enemy populations or hunting efficiency directly (Root 1973, Landis et al.
2000, Sunderland and Samu 2000).
These complex habitats can also indirectly influence arthropod abundance. Complex
habitats can support greater herbivore species richness, enhancing natural enemy populations
(Andow 1988). These same habitats can also influence hunting behavior of predators,
influencing predation and herbivore populations (Andow 1988). Additionally, natural enemies
rely on visual and odor cues when choosing habitats and locating herbivores (Adedipe and Park
2010, Randkofer et al. 2011). Vegetation and structural complexity may reduce herbivore
populations if cues from these habitats are attractive to natural enemies at higher trophic levels,

10

leading to greater biological control of herbivores (Kaplan 2012). Alternatively, habitat
complexity may interfere with hunting and reduce efficiency of natural enemies through the
interference of visual or odor cues, enhancing herbivore populations at lower trophic levels
(Randkofer et al. 2011, Harvey and Fortuna 2012).
In summary, complex habitats may directly influence the abundance of an arthropod
through changes in the environment or the arthropod’s behavior (Figure 1.3) or indirectly
influence arthropods through changes in abundance or efficiency of arthropods at higher or lower
trophic levels (Figure 1.4).

11

d

d

c

c

b

b

a

a

Vegetational complexity
(living)

Structural complexity
(non-living)

Figure 1.3. Direct impacts of habitat complexity on trophic levels in cabbage fields. Direct
effects include a) Influencing cabbage yield and quality, b) Influencing pest abundance
positively (e.g. provide shelter, protection, and resources or negatively (e.g. interferes with
landing and movement of pest), c) Influencing natural enemy abundance positively (e.g. provide
shelter, protection, and resources) or negatively (e.g. interferes with search efficiency or
chemical and visual cues), and d) Influencing hyperparasitoid abundance positively (e.g. provide
shelter, protection, and resources) or negatively (e.g. interferes with chemical or visual cues
when searching for hosts)

12

d

b

d

b

b,d

b,d

a

c

Vegetational complexity
(living)

Structural complexity
(non-living)

Figure 1.4. Indirect impacts of habitat complexity on trophic levels in cabbage fields. For
instance, a) habitat complexity can change cabbage size and quality influencing pest populations,
b) habitat complexity can change arthropod abundance and location in the field, influencing
habitat choices of higher trophic levels. Changes at higher trophic levels influence lower trophic
levels. For instance, c) habitat complexity affects pest abundance which can influence cabbage
yield and quality or d) habitat complexity can change predator and parasitoid abundance or
efficiency influencing predation and parasitism of lower trophic levels.

13

Thesis objectives. Ultimately the goal of this thesis was to examine alternative management
strategies for pests such as the imported cabbageworm. The first two objectives of this thesis
examined the impact of vegetational and structural habitat complexity on arthropods and
biological control in Michigan cabbage fields. The first objective examined the impacts of
habitat management strategies such as reduced herbicide intensity, strip tillage, and cover crop
mulching on arthropod community structures and abundance. The second objective examined the
impact of different mulch species and the height of the mulch on the arthropod community and
biological control using a field-based technique. The final objective of this thesis was to begin to
determine the mechanisms responsible for trends in natural enemy abundance observed in the
field. In particular, the role of odor cues from mulch on the behavior of natural enemies was
examined.

14

CHAPTER 2.
Cover crop mulch and weed management influence
arthropod communities in strip-tilled cabbage

Introduction. Habitat management techniques that change structural and vegetational
complexity in agroecosystems may affect the density of herbivores, conserve or enhance natural
enemy communities, and affect predation rates of pests. Habitat complexity has been defined
based on measures of non-living and living structural complexity (Rypstra et al. 1999, Finke and
Denno 2002). This includes architecture and/or size of non-living structures (e.g. height or
biomass of dead plant material) as well as measures of living vegetational complexity including
plant species diversity, height, and biomass (McNett and Rypstra 2000, Shrewsbury and Raupp
2000, Obermaier et al. 2008). Agricultural practices typically manipulate different types of
complexity, primarily non-living structural complexity and living vegetational complexity. A
meta-analysis performed by Langellotto and Denno (2004) found that habitat complexity in
agricultural fields increases natural enemy abundance, which can lead to greater biological
control of insect pests. Several mechanisms have been proposed for this enhancement of the
natural enemy community. In agroecosystems, habitats with high non-living structural
complexity have been shown to provide refuge (Kawashima and Jung 2010) and vegetational
complexity can provide resources for natural enemies such as shelter, pollen, and alternative prey
(Root 1973, Landis et al. 2000, Sunderland and Samu 2000). Intraguild predation can also be
reduced as structural and vegetational complexity is increased, enhancing biological control in
greenhouses, aquatic environments, natural terrestrial habitats, and agricultural fields (Janssen et
al. 2007). Habitat complexity appears to influence biological control in agricultural systems,

15

particularly in Brassica crops. Increasing vegetational complexity in Chinese cabbage, broccoli,
and Brussels sprouts with intercropping and living mulches increased natural enemy abundance
and richness (Cai et al. 2007, Xu et al. 2011) and reduced lepidopteran and aphid pests (Costello
and Altieri 1995, Vidal 1997, Hooks and Johnson 2006, Cai et al. 2007, Broad et al. 2008, Xu et
al. 2011).
Other studies have demonstrated that habitat complexity negatively impacts natural
enemy abundance and efficiency. For instance, structurally complex agricultural habitats
negatively impacted hunting behavior or limited the ability of predators in agricultural fields,
such as lady beetles, to visually search for prey (Legrand and Barbosa 2003). Abundance of
Coleomegilla maculata (DeGeer) (Coleoptera: Coccinellidae), a generalist predator in corn, was
also reduced in polycultures compared to corn monocultures (Andow and Risch 1985). Studies
have demonstrated that vegetational complexity can reduce parasitoid abundance or parasitism
rates through interference of visual or chemical signals from hosts in broccoli and Brussels
sprouts (Costello and Altieri 1995, Smid et al. 2002). A complex habitat may change the quality
of these signals, altering or reducing parasitism rates.
Managing an insect community using behavior-modifying tools such as habitat
modification of agroecosystems is challenging because the goal is to enhance crop yields, while
reducing pests and increasing natural enemies. Rather than examining insects according to
taxonomic classifications, classifying insects according to functional traits may be more
informative and allow scientists to find links between groups of pests and natural enemies with
changes in the habitat, allowing the conscious design of habitat management strategies.
Functional traits, such as functional feeding group, feeding mode, diet breadth, and hunting
mode can indicate how groups of insects within a community might respond to specific changes

16

in the environment, or influence predation rates and herbivore abundance (Andow 1991, Harmon
et al. 2003, Schmitz 2009, Szendrei and Rodriguez-Saona 2010, Woodcock and Heard 2011).
This is important because generalist predators and specialized parasitoids may respond to
changes in habitat structure in opposite manners. For example, Tscharntke et al. (2008)
demonstrated that although generalist natural enemy diversity was preserved in complex
landscapes, specialized natural enemies were not, due to specific habitat or resource needs.
Growers can increase the structural and vegetational complexity in their fields through a
variety of methods including intercropping (Liebman and Dyck 1993), interseeding cover crops
into cash crops (Brainard et al. 2004), or planting cash crops into pre-established cover crops
(Nicholson and Wien 1983). Cover crops may act as dead mulches if killed prior to planting, or
may act as “living mulches” if allowed to grow during part or all of the crop life cycle. In striptillage systems, living or dead cover crop mulches can be retained between crop rows while inrow areas are tilled to limit mulch interference with the crop and improve crop establishment
(Luna et al. 2012). Strip-tillage in combination with cover cropping can reduce agrichemical
runoff or ground water contamination, improve soil quality, protect crops and soils from extreme
wind and rain events, and reduce labor and fuel inputs, ultimately improving profitability while
protecting the environment (Luna and Staben 2002, Potter et al. 2008, Mulvaney et al. 2011,
Brainard and Noyes 2012, Luna et al. 2012).
In vegetable crops, both living and dead mulch systems have shown beneficial effects for
reducing insect pests (e.g. Altieri et al. 1985) and suppressing weeds (Liebman and Dyck 1993,
Teasdale 1998). For example, in cereal crops, predation rates of major pests were highest in the
presence of cover crop mulch (Schmidt et al. 2004, Lundgren and Fergen 2010). The impact of
cover crop mulches on arthropod communities in cabbage (Brassica oleraceae L.) is not well

17

understood, although the impact of cover crop mulches on pest populations, particularly
Lepidoptera, has been well studied in other Brassica crops such as broccoli (e.g. Mangan et al.
1995).
Although cover crop mulch systems have shown benefits for weed and insect suppression,
these benefits have often come at the expense of crop yields, due to competition for nutrients,
water or light (Teasdale 1998, Brainard and Bellinder 2004). One approach for minimizing
interference between living mulches and crops is to kill mulches before the crop has reached the
“critical weed free period” (CWFP) (Muller-Scharer and Potter 1991). In weed management
studies, the CWFP is defined as the period during which weeds must be removed to avoid yield
loss (e.g. Martin et al. 2001). For transplanted cabbage, CWFP studies (Weaver 1984) suggest
that cover crops must be terminated within 3 to 4 weeks of transplanting to avoid yield losses.
However, CWFP vary considerably with environmental conditions and weed (or cover crop)
species and density (Weaver et al. 1992), so the critical period of removal for pre-established
cover crop mulches is not well established.
Growers using habitat management strategies such as cover crop mulch may also be able
to reduce herbicide inputs in their fields based on weed suppressive effects of cover crop
mulches. Combining cover crop mulches and reduced herbicide inputs can increase the habitat
complexity between rows. In a variety of fruit crops and field corn, increasing non-crop species
was associated with increases in natural enemy abundance and reductions in pest populations
(Rieux et al. 1999, Letourneau et al. 2011). Increased weed diversity also reduced pest
populations in beans (Andow 1990, Andow 1992), and increased generalist predator abundance
in other vegetable crops, such as collards, corn, cauliflower, and tomato (Altieri et al. 1985,
Schellhorn and Sork 1997). Predation of a common herbivore in beans also increased in weedy

18

plots (Andow 1990). However, weed diversity had little impact on predation by generalist
predators in field corn (Wilson et al. 2004) and reduced specialized natural enemy populations in
canola (Broatch et al. 2010). In addition, beneficial effects of weeds on arthropod communities
must be weighed against potential negative effects on crop quality and yield that are often
associated with low herbicide intensity.
Typical pests in cabbage fields in North America consist of specialized Brassica feeders,
including the cabbage aphid, Brevicoryne brassicae L. (Hemiptera: Aphididae), imported
cabbageworm, Pieris rapae L. (Lepidoptera: Pieridae), and diamondback moth, Plutella
xylostella L. (Lepidoptera: Plutellidae) (Root 1973). An oligophagous cabbage pest, the crucifer
flea beetle, Phyllotreta cruciferae (Goeze)(Coleoptera: Chrysomelidae) and two generalist
herbivores, the cabbage looper, Trichoplusia ni (Hübner) (Lepidoptera: Noctuidae), and onion
thrips, Thrips tabaci (Lindeman)(Thysanoptera: Thripidae), are also of economic importance
throughout the United States (Hines and Hutchinson 2001, Shelton et al. 2008). Specialized
natural enemies, such as the parasitic wasps Cotesia rubecula (Marshall) (Hymenoptera:
Braconidae) and Diadegma insulare (Cresson)(Hymenoptera: Ichneumonidae), and generalist
predators (e.g. spiders, carabids, harvestmen, lady beetles, and predatory stinkbugs) attack these
pests. The natural enemy community in Michigan cabbage fields consists of the spined soldier
bug (Podisus maculiventris (Say), Hemiptera: Pentatomidae), minute pirate bug (Orius sp.,
Hemiptera: Anthocoridae), and predatory thrips (Aelothrips sp., Thysanoptera: Aeolothripidae).
Lady beetle species are dominant predators, primarily the pink lady beetle (Coleomegilla
maculata (DeGeer), Coleoptera: Coccinellidae), fourteen spotted lady beetle (Propylaea
quatuordecimpunctata (Linnaeus), Coleoptera: Coccinellidae), multicolored Asian lady beetle
(Harmonia axyridis (Pallas), Coleoptera Coccinellidae), and convergent lady beetle

19

(Hippodamia convergens (Guerin), Coleoptera: Coccinellidae).
Habitat management techniques, such as flowering strips (Pfiffner et al. 2009) and
companion plants (Hooks and Johnson 2003) have been studied to enhance biological control of
cabbage pests and enhance populations of these natural enemies. However, it is unknown how
levels of mulch and herbicide intensity may influence community structure and abundance of
herbivores and natural enemies in this system.
The primary objective of this study was to examine the impact of increasing both
components of habitat complexity: structural and vegetational complexity between rows, on
herbivore and natural enemy communities and crop yield in cabbage. An oat cover crop killed at
different dates in the spring was used to manipulate the amount of physical structure between
rows—the structural complexity within the habitat. Earlier kill dates translated to less physical
structure as the oat cover crop was smaller when it was terminated, while, at the latest kill date,
oats were much taller and created more structural complexity. The use of high or low herbicide
intensity created different levels of weed biomass, or vegetational complexity. For the purpose of
this study, structural complexity refers to the amount of oat mulch in the row middle, as
determined by oat kill date. Vegetational complexity refers to the amount of weed biomass,
manipulated by herbicide intensity. We determined whether higher structural and vegetational
complexity influenced arthropod community structure, functional groups, and species abundance.
The secondary objective was to assess the impact of oat kill date and herbicide intensity on weed
suppression.

20

Materials and Methods
Experimental Design. I performed a field study at Michigan State University’s Horticulture
Farm in East Lansing, Michigan. In 2010 and 2011, separate fields were used, both with Marlette
fine sandy loam soil. The experimental design in both years was a randomized complete block
design with six treatments and four replications. Treatments were a combination of two main
effects: oat kill date (early, intermediate, or late) and herbicide intensity (low or high). Low
herbicide intensity treatments received standard cabbage pre-emergence herbicides. High
herbicide intensity treatments received pre-emergence herbicides and either an additional postemergence herbicide (2010) or supplemental hand weeding (2011) to effectively exclude all
weeds. In 2010, plots were 3.0 x 7.6 m, with 4 rows of cabbage spaced 76 cm apart. In 2011, plot
size was increased to 4.6 x 9.1 m with 6 rows of cabbage per plot.

Field Operations. The dates of major field operations are provided in Table 2.1. In mid-April
the entire experimental area was fertilized with 91 kg/ha of 19-19-19 (NPK), and oats (Avena
sativa var. “Ida”; Zmitko Farm, Owosso, MI) were planted using a grain drill (Great Plains
Compact Drill 3P606NT, Salina, Kansas). Oats were sown at 126 kg/ha in 2010, and at 175
kg/ha in 2011. In late May, an additional 228 kg/ha of 19-19-19 (NPK) fertilizer was applied
over the entire experimental area in accordance with soil tests and nutrient management
recommendations for cabbage (Warncke et al. 2004). A Hiniker® Model 6000 two-row striptiller (equipped with notched trash-cleaning discs, cutting-coulter, shank-point assembly,
berming disks and rolling basket) was used to create 25 cm x 25 cm deep strips at 76 cm
between-strip spacing. Pre-established oats were effectively killed in the strips, but minimally
affected in the undisturbed 51 cm between strips, thus creating cover crop mulch between crop

21

rows. Following strip-tillage, a pre-transplant herbicide application containing Goal (0.28 kg
active ingredient oxyfluorfen / ha) and Dual II Magnum (0.54 kg active ingredient s-metolachlor
/ ha) was applied in all treatments. For early oat kill-date treatments, Select 2EC (0.21 kg active
ingredient clethodim / ha) was also applied at this time. Intermediate oat kill treatments with
Select EC occurred 14 days after transplanting (DAT) in 2010, and 9 DAT in 2011. Late oat kill
treatments occurred 27 DAT in 2010 and 21 DAT in 2011. In 2010, high herbicide intensity plots
received a post-emergence herbicide application to manage broadleaf weeds 28 DAT (Goal, 0.28
kg active ingredient oxyfluorfen / ha); these plots were also hand-weeded 31 DAT. Due to
reduced weed pressure in 2011, no additional post-emergence herbicides were applied in high
herbicide intensity plots, though plots were hand-weeded at 29 and 71 DAT to remove escaped
weeds.
Cabbage transplants (var. “Blue Dynasty”; Siegers Seed Company, Holland, MI) were
grown to the 4-6 leaf stage in 98 cell plug trays and hardened off for either 4 d (2010) or 14 d
(2011) prior to transplanting in the field. Cabbage was transplanted by hand on May 21, 2010
and May 31, 2011 into strip-tilled zones (76 cm between rows) with 36 cm between transplants
within the row. No fungicides were used. In 2011, an insecticide selective for lepidopteran pests,
Intrepid (0.56 kg active ingredient methoxyfenozide / ha), was applied when imported
cabbageworm damage became severe 43 DAT. Cabbage was side-dressed with urea at 45 kg
N/ha on 22 June 2010 and 27 June 2011.

22

Table 2.1. Dates of major field operations in 2010 and 2011.
Date
Field operation
2010
2011
Oats planted
14-Apr
14-Apr
Strip-tillage
20-May
24-May
Oxyfluorfen/Dual II Magnum applied
20-May
24-May
Clethodim applied in early oat kill
20-May
24-May
Cabbage transplanted
21-May
31-May
Clethodim applied in intermediate oat kill
14-Jun
9-Jun
Clethodim applied in late oat kill
17-Jun
21-Jun
Oxyflourfen applied in high herbicide
intensity
11 June
NA
Urea sidedress application
22-Jun
27-Jun
Hand weeding
21-Jun
29-Jun
Hand weeding 2 in high herbicide intensity
NA
10-Aug
Cabbage harvest
17-Aug
8-Sep
2

Oat height and biomass. In 2010, I measured oat height in two 0.25 m quadrats randomly
2

placed in each plot. The heights of ten random oat plants per 0.25 m quadrat were measured
prior to cabbage transplant and 10, 26, and 52 DAT in all plots. During the final height
measurements, one quadrat of dead oat biomass was collected per plot to determine oat dry
weight. Above ground oat biomass was dried in the oven for 7 d at 60°C prior to weighing. In
2011, oat height was measured using the same methods as 2010. However, oat height was
recorded in intermediate, late, and all treatments 9, 16, and 46 DAT respectively. Killed oat
2

biomass in 2011 was collected from one 0.25 m quadrat per plot at 9 and 17 DAT, in
intermediate and late kill-date plots respectively. Killed oat biomass was collected 0 and 3 d
prior to oat kill-date in intermediate and late killed plots. Oats were not present at these dates in
early kill date plots.

23

Weed biomass. A weed census was conducted at the end of the cabbage growing season in 2010,
84 DAT. Two weed species, common lambsquarters (Chenopodium album L.) and smartweed
2

(Polygonum pensylvanicum L.), dominated the weed community. In two 0.25 m quadrats per
plot, lambsquarter and smartweed biomass was collected, dried, and weighed as described
previously for oats. In 2011, a weed census was conducted 45, 84, and 90 DAT. Compared to
2010, weed density was lower in low herbicide intensity treatments, but a more diverse weed
community was present including, in order of abundance: common lambsquarters, common
ragweed (Ambrosia artemisiifolia L.), fall panicum (Panicum dichotomiflorum Michx) and wild
radish (Raphanus raphinistrum L.). Weed biomass was collected for common lambsquarters and
2

common ragweed (>95% of total weed biomass) from four 0.25 m quadrats in each plot. A
larger sampling area was used in 2011 because weed density was lower in 2011 compared to
2010.

Cabbage biomass and yield. To determine mid-season cabbage plant biomass, I randomly
collected four cabbage plants per plot 40-46 DAT in 2010 and 27 DAT in 2011. Cabbage plants
were cut above the root crown, washed, and weighed. The numbers of leaves per plant were
counted and total leaf area was measured (Portable Leaf Area Meter LI-3000C 1.0.0, LI-COR
Biosciences, Lincoln, Nebraska). For final cabbage yield, on 17 August 2010, all cabbage heads
were harvested from the middle 7 m sections of the two center rows for each cabbage plot. Final
cabbage yield in 2011 was taken on 8 September 2011 from the middle 4.6 m sections of the
center four rows of each plot. In both years, the number and weight of marketable and nonmarketable heads were assessed, with those greater than 11.4 cm in diameter classified as
marketable.

24

Habitat and Cabbage Yield Statistical Analysis. The effects of oat kill date and herbicide
intensity on oat height, oat biomass, weed biomass, cabbage biomass (marketable and nonmarketable), cabbage leaf area, and number of cabbage leaves per plant were assessed using a
mixed model (PROC Mixed, SAS 9.2, 2002-2008) with herbicide intensity and oat kill date as
fixed factors, and block as a random factor. Oat height was averaged across three sampling dates.
Oat biomass and weed biomass were summed for each plot. Oat biomass was log transformed (x
+ 0.5) to meet assumptions of normality and equal variance. The other dependent variables were
tested and transformations were not necessary. A least-square means test was performed if main
effects or their interactions were significant (at α=0.05). For each dependent variable, the two
years were analyzed separately.

Insect community assessment. Insects were sampled visually and with yellow sticky traps (7.5
× 12.3 cm, Great Lakes IPM, Vestaburg, MI) in the growing season for 9 wks in 2010 and 14
wks in 2011. Sessile insects were evaluated visually once a week on 10 randomly sampled
cabbage plants per plot in the center cabbage rows. All cabbage leaves were examined and
insects on them were identified and counted. Mobile or flying insects were sampled using yellow
sticky traps placed 0.4 m from ground level on metal stakes. Traps were changed weekly and
insects were identified to species or to the lowest possible taxonomic classification (Marshall
2006). Voucher specimens for the imported cabbageworm, diamondback moth, and
representatives of the natural enemy community were deposited in the A.J. Cook Arthropod
Research Collection at Michigan State University (East Lansing, MI).

25

To determine the impact of treatments on insect community structure, cluster analysis
(PROC Cluster) was used based on average abundance per treatment of each insect species. The
analysis was repeated using average abundance per plot for each insect species. Herbivores and
natural enemies were analyzed separately in both years. This type of cluster analysis uses
average linkage distance to determine similarity between groups (Proc Cluster).
To determine the impact of treatments on individual species, a repeated measures
analysis was used. Due to the difference in environmental conditions between 2010 and 2011
(e.g. precipitation, degree day accumulation), data from visual sampling and sticky traps was
transformed using log (x + 0.01) and analyzed separately for 2010 and 2011. Analysis included
date as a repeated factor, block as a random factor, and treatment nested within block as the
subject. Fixed factors included herbicide intensity, oat kill date, and interaction between
herbicide intensity and oat kill date (PROC Mixed). Covariates included marketable cabbage
yield, total herbivore abundance, total predator abundance, total parasitoid abundance, and host
abundance. For instance, diamondback moth abundance was included as a covariate for
analyzing Diadegma insulare abundance and imported cabbageworm abundance was included as
a covariate for Cotesia rubecula abundance. If main effects were significant (α=0.05), we
performed least-square means tests with Tukey-Kramer adjustment.
In order to identify how the herbicide intensity and oat kill date treatments influenced
functional groups of natural enemies and herbivores, Hedge’s d effect size (Cohen 1988) was
calculated using Metawin 2.0 (Sinaur Associates, Sunderland, MA). A positive or negative effect
size indicates a positive or negative impact of habitat complexity on insect abundance. Any
confidence interval overlapping with zero indicates a lack of significant difference between the
lowest and highest complexity habitats for that functional group. Insect species from visual and

26

sticky trap sampling were categorized as herbivore, predator, or parasitoid, and according to
feeding guild (chewing, sucking, none), hunting mode (active or sit and wait), and diet breadth
(specialist, generalist, oligophagous) (specialists: herbivores with hosts within a single plant
family, oligophagous: herbivores with hosts in two plant families, generalists: herbivores with
hosts in more than two plant families)(Supp. Table S1). Effect size was used to compare the
effects of the two extreme treatments (high herbicide intensity and early oat kill date, compared
to low herbicide intensity and late oat kill date) (Cohen 1988, Horton et al. 2003). These two
extreme treatments will be referred to as the lowest habitat complexity and highest habitat
complexity treatments from here on. For less mobile herbivores that often concentrated on their
food source, visually collected insect abundance data from these two treatments was used to
calculate average herbivore species’ abundance per plot for each week. Sticky trap data from
these two treatments were used to calculate average natural enemy species abundance and
mobile herbivore abundance per plot per week. A fixed effects model was used with categorical
structure and included re-sampling tests (999 iterations) that calculate cumulative effect size with
95% confidence intervals. Mean effect size (d+) and 95% bootstrap confidence intervals were
calculated for each feeding category (Gurevitch and Hedges 1999). Effect sizes of 0.2, 0.5. and
0.8 are considered small, intermediate, and large respectively (Cohen 1988). To determine
significance, between-group heterogeneity (Qb) was considered with a chi-square distribution.

Results
Oat height and biomass. In 2010, later oat kill date, weed management, and the interaction
between these factors positively influenced oat height (Table 2.2). Within low and high herbicide
intensity plots, oats were more than three times taller in late-killed oats, compared to the

27

intermediate-killed plots (t > 27.35, df=15, P < 0.01). In 2011, oat height did not differ across
plots with different herbicide intensity, but plots with a late kill date had 10% taller oats than the
intermediate kill date plots (t = 3.15, df=18, P < 0.01).
In both years, oat biomass was influenced by oat kill date but not herbicide intensity
2

(Table 2.2). In 2010, late killed oats produced 173.63 g/ m of biomass after termination, while
early killed oat plots did not contain substantial oat biomass on sampling dates, since these
decayed before the sampling date. In 2011, three times more biomass was produced in late killed
oats than intermediate killed oats (t = 3.15, df=15, P < 0.01). Both late and intermediate killed
oats produced greater biomass than early killed oats (t > 26.77, df=15, P < 0.01).

Weed biomass. In both years, herbicide intensity but not by oat kill date influenced weed
2

biomass(Table 2.2). In 2010, low herbicide intensity plots produced 266 g/ m biomass
2

compared to only 2 g/m in high herbicide intensity plots (t = 6.24, df=15, P < 0.01). In 2011,
weed biomass was far lower than in 2010, with greater biomass in low- compared to highintensity herbicide plots. We had anticipated that oat mulch might provide weed suppressive
benefits through shading or competition for water or nutrients. However, no such effect was
detected, likely due in part to large within-treatment variability in weed biomass.

Cabbage biomass and yield. Oat kill date but not herbicide intensity influenced mid-season
cabbage biomass and leaf area per plant in 2010 (Supp. Table S2). Cabbage leaf area in early and
intermediate kill-date plots did not differ, but leaves were 36% greater in late kill-date plots.
Early and intermediate kill date plots also had 36 - 64% greater mid-season cabbage biomass

28

than late kill-date plots. While the leaf area per cabbage plant differed across oat kill date, the
number of cabbage leaves per plant did not (F = 0.84, df=2,15, P > 0.45). In 2011, mid-season
cabbage biomass, leaf area, and number of leaves did not differ across oat kill date or herbicide
intensity (F < 1.89, df=2,15, P > 0.19; Supp. Table S2). This discrepancy between years may
have been due in part to the fact that cabbage was sampled at a later date in 2010 compared to
2011, so greater interference from oats may have occurred by the time of sampling. Oat biomass
in late kill date treatments was also lower in 2011 (Table 2.2), due to a slightly earlier kill date
relative to cabbage transplanting.
The final yield (Table 2.3) and number of marketable and non-marketable cabbage (Supp.
Table S3) in 2010, was influenced by oat kill date. Marketable cabbage biomass was two times
greater in early compared to late kill date plots (t = 5.43, df=15, P < 0.01) and 64% greater in
intermediate kill date plots (t = 1.75, df=15, P = 0.1). Early kill date plots had 37% more
marketable heads than late kill date plots (t = 3.39; df=15; P < 0.01) and intermediate kill date
plots had marginally more marketable heads (23%) than late kill date plots (t = 2.10, df=15, P =
0.05). Non-marketable heads were more abundant in late kill date plots compared to plots with
oats killed at an early (t = 3.58, df=15, P < 0.01) or intermediate date (t = 2.21, df=15, P = 0.04).
In 2011, marketable cabbage yield was 57% higher in early compared to late kill date
plots (t = 3.24, df=15, P < 0.01; Table 2.3). Herbicide intensity and oat kill date did not affect
non-marketable yield. Non-marketable biomass and the number of marketable or non-marketable
cabbage heads per plot in 2011 were not different across herbicide intensity (F < 1.66, df =1,15,
P > 0.21; Table 2.3, Supp. Table S3) or oat kill date treatments (F < 2.65, df =2,15, P > 0.10).
Herbicide intensity affected marketable yield and number in 2010, but had no affect on
final yield or number in 2011 (Table 2.3, Supp. Table S3). In 2010, marketable cabbage weight

29

was reduced by 33% under low compared to high herbicide intensity treatments (t = 3.04, P <
0.01). Greater yield loss due to weeds in 2010 was likely the result of greater weed biomass in
2010 compared to 2011 (Table 2.2).

Insect community structure. In 2010, there were two distinct herbivore communities according
to herbicide intensity (Fig. 2.1A). Herbivore communities were more similar between plots
where oats were killed at an intermediate or later date compared to plots where oats were killed
early in the season. In 2011, herbivore communities were similar in treatments where oats were
killed late in the season, irrespective of herbicide intensity. The herbivore communities in plots
with an early or intermediate oat kill date were similar within the plots with low herbicide
intensity.
In 2010, the natural enemy communities were similar in late oat kill date plots regardless
of herbicide intensity (Fig. 2.1B). There were distinct natural enemy communities in early or
intermediate oat kill date plots with the two different herbicide intensities. The natural enemy
communities within early and intermediate kill date treatments were particularly similar across
low herbicide intensity. In 2011, the natural enemy communities were similar in early and
intermediate kill date treatments, but the communities within plots with an intermediate kill date
and low herbicide intensity were more similar to the communities found where oats were killed
later in the season.
Herbivore and natural enemy communities were examined across plots (Supplementary
Figure S1). Although there was variation within treatments, plot clustering supports previous
findings.

30

Insect abundance. Since interactions between main effects were not significant for insect
abundance, only the results for main effects are presented. A couple herbivore and natural enemy
species consistently did not respond to experimental treatments in 2010 or 2011. Abundance of
the diamondback moth were not different among any of the oat kill date (Table 2.4) or herbicide
intensity treatments (Table 2.5). In addition, the abundance of the ichneumonid wasp Diadegma
insulare, a specialist parasitoid that attacks the diamondback moth, was not significantly
different among treatments in either 2010 or 2011.

Impact of oat kill date on insect abundance. Oat kill date had an impact on herbivore
abundance, for example the abundance of the cabbage looper (a pest that migrates to Michigan
annually) tended to have 85% higher abundance in plots with the intermediate compared to early
kill dates in 2010 (t = 2.50, df=19, P = 0.05). In 2011, abundance of the cabbage looper was
more than 60% greater in early kill dates than intermediate or late kill dates (t > 3.42, df= 28, P <
0.01). Crucifer flea beetle abundance was 37% greater in late kill dates than earlier in the 2010
season (t > 3.06, df= 26, P < 0.01). Onion thrips abundance was 36% higher in intermediate
compared to early kill date plots in 2011 (t = 2.84, df= 27, P = 0.02).
Generalist predators such as lady beetle larva and adults typically responded positively to
later oat kill dates (Table 2.6). Lady beetle larvae were three times more abundant where oats
were killed later in the season compared to plots with oats killed at the intermediate date in 2010
(t = 2.58, df=15, P = 0.05). Pink lady beetles were 99% more abundant in late than early kill date
plots (t = 3.07, df=20, P = 0.02). Fourteen spotted lady beetles were three times more abundant
in late than early kill date plots in 2011 (t = 2.56, df=15, P = 0.03) and predatory thrips were
significantly different in 2011, with 25% more in intermediate compared to early oat kill date

31

plots (t = 3.24, df=20, P = 0.01). In contrast, convergent lady beetles were 77% more abundant in
intermediate than late kill date plots in 2011 (t = 2.89, df=24, P = 0.02).
Abundance of one species of parasitic wasp and one hyperparasitoid species responded to
oat kill date (Table 2.6). The braconid wasp, Cotesia rubecula, a parasitoid that attacks the
imported cabbageworm, was 31% more abundant in late kill date plots compared to plots where
oats were killed earlier in the 2011 season (t = 12.72, df = 24, P = 0.03). A group of
hyperparasitoids in the superfamily Chalcidoidea was 59% (t = 3.75, df=24, P < 0.01) and 84%
(t = 4.03, df= 26, P < 0.01) more abundant in plots with oats killed later in the season than plots
with an early kill date, in 2010 and 2011 respectively.

Impact of herbicide intensity on insect abundance. Insect abundance was influenced by
herbicide intensity. In 2010, cabbage loopers were 40% more abundant in high herbicide
intensity plots compared to low herbicide intensity plots in 2010 (t = 2.34, df=19, P = 0.03). In
2010, cabbage aphids were nearly three times more abundant in high compared with low
herbicide intensity plots (t = 8.31, df=23, P < 0.01). In 2011, this trend remained the same, but
aphids were only 28% more abundant (t = 3.15, df=15, P < 0.01). In 2011, 20% more imported
cabbageworm larvae were found in high (t = 2.11, df = 24, P = 0.04; Table 2.5) compared to low
herbicide intensity plots.
Generalist predators, including the spined soldier bug, pink lady beetle, fourteen spotted
lady beetle, and minute pirate bug were impacted by herbicide intensity in at least one year
(Table 2.7). Spined soldier bugs in 2010 were three times more abundant in low than in high
herbicide intensity plots (t = 3.93, df = 19, P < 0.01). Abundance of the pink lady beetle was
70% more abundant in low than high herbicide intensity plots in 2010 (t = 4.26, df=24, P < 0.01).

32

Another species, the fourteen spotted lady beetle, was three times more abundant in low
compared to high herbicide intensity plots in 2010 (t = 4.62, df=15, P < 0.01). The minute pirate
bug, was three times more abundant in low than high herbicide intensity plots in 2010 (t = 4.85,
df=19, P < 0.01) and 84% more abundant in low than high herbicide intensity plots in 2011 (t =
2.04, df=18, P = 0.04).
Although abundance of parasitic wasps at the third trophic level did not differ across
herbicide intensity treatments, a member of the fourth trophic level, in the superfamily
Chalcidoidea, was 13% more abundant in low compared with high herbicide intensity plots in
2010 (t = 2.36, df= 24, P = 0.03; Table 2.7).

Significant covariates for herbivores. Marketable cabbage yield was a significant
covariate for cabbage aphids and crucifer flea beetles in 2010 (F > 5.46, P < 0.02) and imported
cabbageworm in 2011 (F = 10.21, df= 1,24; P = 0.01). In these cases, greater marketable yield
positively impacted cabbage aphid abundance and negatively impacted flea beetle and imported
cabbageworm abundance. Total predator abundance was a significant covariate for the imported
cabbageworm and diamondback moth in 2010 (F > 4.43, P < 0.04) and both cabbage loopers
and onion thrips in 2011 (F > 20.54, P < 0.01). Predator abundance negatively impacted
herbivore abundance, excluding onion thrips which were more abundant in plots with greater
predator abundance.

Significant covariates for natural enemies. Marketable cabbage yield was a significant
covariate for the pink lady beetle in both 2010 and 2011 (F> 8.10, P < 0.01), with fewer lady
beetles in plots with greater marketable yield. Total herbivore abundance also affected pink lady

33

beetle abundance in 2011 (F = 6.84, df= 1,252; P < 0.01), with fewer herbivores in plots with the
greatest number of pink lady beetles. Total predator abundance was a significant covariate for
the pink lady beetle, fourteen spotted lady beetle, convergent lady beetle, lady beetle larvae,
predatory thrips, Cotesia rubecula, Diadegma insulare, and Chalcidoidea hyperparasitoids (F>
9.56, P < 0.02). In general, greater abundance of these natural enemy species was found in plots
with greater total predator abundance. Finally, abundance of the imported cabbageworm and
Cotesia rubecula were significant covariates in at least one year for C. rubecula and its
hyperparasitoids in the superfamily Chalcidoidea respectively (F > 3.87, P < 0.05), with host
abundance positively influencing parasitoid or hyperparasitoid abundance.

Insect functional groups. The effect of habitat complexity on functional groups in cabbage was
examined using Hedge’s d effect size. Any significant heterogeneity indicates significant
differences between the two extreme treatments (e.g. low herbicide intensity with oats killed late
and high herbicide intensity with oats killed early). Positive or negative effect sizes indicate a
positive or negative impact of habitat complexity. We detected significant heterogeneity of effect
sizes in 2010 (QT > 232.28, P < 0.01; n > 180) and in 2011 (QT > 247.36, P < 0.01; n > 184) for
herbivore abundance, and natural enemy abundance. Life stages of insects and location of insects
in the field did not explain heterogeneity in effect sizes in 2010 (QB < 5.26, P > 0.20) or 2011
(QB < 3.64, P > 0.30), (data not shown). However, effect sizes between functional feeding
groups was significant in 2011 (QB = 9.73, P =0.02). Predators were more abundant in high
habitat complexity plots with a significant, small effect size. There was no effect of habitat
complexity on herbivores or parasitoids. Feeding guild did account for effect size variation in

34

2011 (QB = 10.63, P < 0.01) when chewing herbivores were less abundant in high complexity
plots. Significant effect sizes for chewing and non-feeding insects were small in 2011. Effect
sizes for groups with different diet breadths were significant in 2010 (QB = 8.59, P = 0.01) and
in 2011 (QB = 6.09, P = 0.05). Oligophagous herbivores were less abundant in high complexity
plots in 2010 with a significant, moderate effect size. In 2011, specialist feeders were reduced in
high complexity plots. Habitat complexity led to a significant, small effect size for specialist
feeders. Effect sizes for groups of natural enemies with different hunting modes (e.g. active or sit
and wait) were significant in 2011 (QB = 13.84, P < 0.01), although not in 2010 (QB = 2.25, P =
0.52). In 2011, a small and moderate effect size was observed for active and for sit and wait
predator abundance, respectively. Active hunters were significantly more abundant in high
complexity plots, whereas sit and wait predators were more prevalent in low complexity plots.

35

Table 2.2. Mean ± SEM oat height, oat biomass, and weed biomass in 2010 and 2011.
a,b

Main effects
Oat kill date
Early
Intermediate
Late
Herbicide
intensity
Low
High
Significance
Oat kill date
(KD)
Herbicide
intensity (HI)
KD x HI

Oat height (cm)
2010
2011
0.0 ± 0.0a
8.21 ± 0.51b
26.75 ± 0.72c

0.0 ± 0.0a
41.67 ± 0.93b
46.56 ± 1.53c

11.11 ± 3.17a
12.19 ± 3.60b

29.58 ± 6.36
29.24 ± 6.34

2 a,c

Oat biomass (g/ m )
2010
2011

0.0 ± 0.0a
0.0 ± 0.0a
NA
28.21 ± 5.66b
173.63 ± 13.61b 87.06 ± 8.40c

53.67 ± 23.51
62.08 ± 27.22

39.59 ± 12.25
37.26 ± 11.51

2 a

Weed biomass (g/ m )
2010
2011
167.28 ± 83.09
131.21 ± 51.96
104.23 ± 46.68

9.03 ± 5.92
34.15 ± 15.79
10.39 ± 8.47

266.26 ± 44.16a 35.71 ± 10.83a
2.22 ± 1.08b
0.0 ± 0.0b

-------------------------------------------------------P-value----------------------------------------------------------<0.01
<0.01
<0.01
<0.01
0.49
0.06
<0.01

0.79

0.39

0.74

<0.01

<0.01

<0.01

0.77

0.47

0.84

0.49

0.06

a
Different letters within column indicate significant differences within main effects according to LSMeans (α=0.05).
b
These represent average heights during the 2010 or 2011 season.
c

These represent maximum weights during the 2010 or 2011 season.

36

Table 2.3. Mean ± SEM non-marketable and marketable fresh weight of cabbage heads in 2010 and
2011.
a

Main effects
Oat kill date
Early
Intermediate
Late
Herbicide intensity
Low
High
Significance
Oat kill date (KD)
Herbicide intensity
(HI)
KD x HI

Marketable (kg)
2010
2011

a

Non-marketable (kg)
2010
2011

48.51± 3.54a
40.86± 4.20a
24.84± 4.44b

60.31± 6.37a
50.20± 7.28ab
38.51± 4.27b

0.36± 0.14a
1.83± 0.77ab
2.86± 0.74b

2.37± 0.69
3.56± 1.29
4.03± 0.82

32.65± 4.11a
43.49± 4.08b

46.09± 5.29
53.25± 5.61

1.36± 0.49
2.01± 0.65

3.73± 0.68
2.91± 0.89

--------------------------------------P-value---------------------------------------<0.01
0.02
0.03
0.38
<0.01
0.21
0.35
0.42
0.70

0.83

a

0.83

Different letters within column indicate significant differences within main effects according to
LSMeans (α=0.05).

37

0.65

Table 2.4. Mean ± SEM herbivore abundance in experimental plots across fixed effect: oat kill date (early,
intermediate, or late) in 2010 and 2011.
b
Source of variation
Insect abundance (Mean ± SEM)
Early oat kill
date plots
2010 Major Herbivores
c,f
Imported cabbageworm
Diamondback moth
Cabbage looper

c,f

c

Late oat kill
date plots

Oat kill date
Df

a

F

P

5.45 ± 0.36

9.19 ± 0.87

8.48 ± 1.55

2,20

2.26

0.13

16.32 ± 1.16

14.97 ± 1.01

16.81 ± 2.74

2,21

0.79

0.47

2,19

4.72

0.02

4.14 ± 0.50a

d,e

Intermediate oat
kill date plots

7.64 ± 1.01b

7.38 ± 1.65ab

37.08 ± 2.86

35.16 ± 3.04

27.66 ± 2.60

2,23

0.78

0.47

18.60 ± 1.64a

17.19 ± 1.61a

23.51 ± 2.18b

2,25

6.34

<0.01

d

55.00 ± 3.59

67.32 ± 5.55

49.21 ± 3.18

2,15

1.40

0.28

2011 Major Herbivores
c,e
Imported cabbageworm

12.53 ± 1.00

13.00 ± 1.00

13.02 ± 1.13

2,24

0.29

0.75

2.82 ± 0.25

2.23 ± 0.21

2.09 ± 0.21

2,18

2.60

0.10

0.08 ± 0.03a

0.05 ± 0.03b

0.04 ± 0.02b

2,26

7.66

<.01

63.56 ± 8.78

63.44 ± 12.08

72.50 ± 11.27

2,15

1.60

0.23

76.53 ± 10.38

77.81 ± 8.54

75.10 ± 14.05

2,15

1.44

0.27

50.84 ± 3.12a

79.81 ± 5.68b

73.95 ± 4.98ab

2,26

4.40

0.02

Cabbage aphid

Crucifer flea beetle
Onion thrips

d,e

Diamondback moth
Cabbage looper
Cabbage aphid

c,f

d

Crucifer flea beetle
Onion thrips

c

d,f

d

a
Denominator degrees of freedom were calculated with the Kenward-Roger method (SAS 9.2, 2002-2008).
b
Different letters indicate significant differences within row according to LSMeans (α=0.05).
c
Abundance determined from visual abundance per plot.
d
Abundance determined from sticky trap abundance per plot.
e
Abundance determined using model with marketable cabbage yield as a covariate.
f

Abundance determined using model with total predator abundance as a covariate.

38

Table 2.5. Mean ± SEM herbivore abundance in experimental plots across fixed effect: herbicide intensity (low or
high) in 2010 and 2011.
b
Source of variation
Insect abundance (Mean ± SEM)
Low herbicide
intensity plots
2010 Major Herbivores
c,f
Imported cabbageworm
Diamondback moth

c,f

High herbicide
intensity plots

Herbicide intensity
df

a

F

P

7.29 ± 0.59

8.10 ± 1.06

1,20

0.46

0.51

15.49 ± 1.15

16.59 ± 1.87

1,22

1.26

0.27

c

5.35 ± 0.69a

7.41 ± 1.13b

1,19

5.88

0.03

d,e

17.34 ± 1.15a

49.05 ± 2.67b

1,23

68.98

<0.01

16.61 ± 1.15

22.81 ± 1.74

1,25

0.62

0.44

d

56.68 ± 3.99

57.83 ± 2.96

1,15

0.18

0.67

2011 Major Herbivores
c,e
Imported cabbageworm

11.70 ± 0.79a

14.00 ± 0.91b

1,24

4.44

0.04

2.36 ± 0.19

2.40 ± 0.18

1,18

1.56

0.23

0.09 ± 0.03

0.02 ± 0.01

1,27

0.65

0.43

58.23 ± 6.16a

74.77 ± 9.18b

1,15

9.93

<0.01

70.61 ± 9.85

82.36 ± 10.50

1,32

0.39

0.54

70.92 ± 3.93

65.48 ± 3.90

1,27

0.27

0.61

Cabbage looper
Cabbage aphid

Crucifer flea beetle
Onion thrips

d,e

Diamondback moth
Cabbage looper
Cabbage aphid

c,f

d

Crucifer flea beetle
Onion thrips

c

d,f

d

a
Denominator degrees of freedom were calculated with the Kenward-Roger method (SAS 9.2, 2002-2008).
b
Different letters indicate significant differences within row according to LSMeans (α=0.05).
c
Abundance determined from visual abundance per plot.
d
Abundance determined from sticky trap abundance per plot.
e
Abundance determined using model with marketable cabbage yield as a covariate.
f

Abundance determined using model with total predator abundance as a covariate.

39

Table 2.6. Mean ± SEM natural enemy abundance in experimental plots across fixed effect: oat kill date (early, intermediate, or
late) in 2010 and 2011.
b
Source of variation
Insect abundance (Mean ± SEM) in oat kill date plots
Early
Intermediate
Late
Oat kill date
a
2010 Major Natural Enemies
F
P
df
c
1.20 ± 0.44
2.71 ± 0.69
1.36 ± 0.28
2,19
1.20
0.32
Spined soldier bug
d,e
0.97 ± 0.11
0.78 ± 0.09
1.30 ± 0.13
2,25
0.15
0.87
Pink lady beetle
d
0.46 ± 0.07
0.67 ± 0.09
0.76 ± 0.09
2,15
1.39
0.28
14 spotted lady beetle
c
0.34 ± 0.08a
0.33 ± 0.09a
0.96 ± 0.17b
2,15
4.26
0.03
Lady beetle larvae
d
5.40 ± 0.63
4.27 ± 0.51
3.21 ± 0.41
2,19
1.74
0.20
Minute pirate bug
d
18.03 ± 1.50
19.34 ± 1.91
11.14 ± 1.15
2,15
1.24
0.32
Predatory thrips
d
10.78 ± 1.44
6.30 ± 0.86
6.51 ± 0.92
2,24
1.33
0.28
Cotesia rubecula
d,i
23.78 ± 1.13a
28.37 ± 1.23a
37.82 ± 1.44b
2,24
7.75
< 0.01
Chalcidoidea
2011 Major Natural Enemies
c
0.64 ± 0.21
0.45 ± 0.17
0.63 ± 0.27
2,19
0.05
0.95
Spined soldier bug
d,e,f,g
0.76 ±0.08a
1.20 ± 0.12ab
1.51 ± 0.13b
2,19
5.04
0.02
Pink lady beetle
d,g
0.05 ± 0.01
0.14 ± 0.03
0.22 ± 0.04
2,12
2.43
0.11
14 spotted lady beetle
d,g
0.11 ± 0.02ab
0.39 ± 0.06a
0.09 ± 0.02b
2,25
4.18
0.03
Convergent lady beetle
c,g
0.08 ± 0.02
0.19 ± 0.04
0.38 ± 0.08
2,20
3.19
0.06
Lady beetle larvae
d
0.84 ± 0.10
1.63 ± 0.21
1.50 ± 0.16
2,18
1.78
0.20
Minute pirate bug
d,g
2.48 ± 0.23a
3.10 ± 0.30b
3.06 ± 0.28ab
2,20
5.40
0.01
Predatory thrips
d,e,h
1.30 ± 0.18a
1.32 ± 0.19ab
1.70 ± 0.28b
2,24
4.00
0.03
Cotesia rubecula
d,e
4.02 ± 0.66
2.48 ± 0.50
2.42 ± 0.54
2,25
1.26
0.25
Diadegma insulare
d,g,i
17.31 ± 0.77a
21.33 ± 1.03a
31.88 ± 2.31b
2,25
8.62
<0.01
Chalcidoidea
a
Denominator degrees of freedom were calculated with the Kenward-Roger method (SAS 9.2, 2002-2008).
b
Different letters indicate significant differences within row according to LSMeans (α=0.05).
c
Abundance determined from visual abundance per plot.
d

Abundance determined from sticky trap abundance per plot.

40

Table 2.6 (cont’d)
e
Abundance determined using model with marketable cabbage yield as a covariate.
f
Abundance determined using model with total herbivore abundance as a covariate.
g
Abundance determined using model with total predator abundance as a covariate.
h
Abundance determined using model with imported cabbageworm abundance as a covariate.
i

Abundance determined using model with Cotesia rubecula abundance as a covariate.

41

Table 2.7. Mean ± SEM natural enemy abundance in experimental plots across fixed effect:
herbicide intensity (low or high) in 2010 and 2011.
b
Source of variation
Insect abundance (Mean ± SEM)
Low herbicide
intensity plots
2010 Major Natural
Enemies
c
Spined soldier bug
d,e
Pink lady beetle
d
14 spotted lady beetle
c

Lady beetle larvae
d
Minute pirate bug
d
Predatory thrips
Cotesia rubecula
d,i
Chalcidoidea

2011 Major Natural
Enemies
c
Spined soldier bug
d,e,f,g
Pink lady beetle
d,g

14 spotted lady beetle
d,g
Convergent lady beetle
c,g
Lady beetle larvae
d

Minute pirate bug
d,g
Predatory thrips
d,e,h
Cotesia rubecula
d,e
Diadegma insulare
d,g,i
Chalcidoidea

Herbicide intensity
a

2.61 ± 0.52a

0.88 ± 0.24b

df
1,19

1.28 ± 0.11a
0.93 ± 0.08a
0.59 ± 0.11
6.44 ± 0.53a

0.75 ± 0.08b
0.33 ± 0.05b
0.48 ± 0.09
2.20 ± 0.23b

1,24
1,15
1,15
1,19

18.14
21.34
0.84
23.49

<0.01
<0.01
0.37
<0.01

18.39 ± 1.53

1,15

0.02

0.88

9.06 ± 1.07
28.17 ± 1.15b

1,24
1,24

0.71
5.58

0.41
0.03

0.65 ± 0.17
1.07 ± 0.08

1,19
1,20

0.65
0.11

0.43
0.75

0.14 ± 0.03
0.17 ± 0.03
0.31 ± 0.06
0.93 ± 0.08b

1,22
1,25
1,21
1,18

1.43
0.29
4.09
6.24

0.24
0.59
0.06
0.02

1,20
1,25
1,26
1,26

0.13
4.06
2.98
0.77

0.72
0.06
0.10
0.39

13.98 ± 0.97
6.64 ± 0.69
31.75 ± 1.05a

d

High herbicide
intensity plots

0.50 ± 0.18
1.25 ± 0.10
0.13 ± 0.02
0.23 ± 0.04
0.12 ± 0.02
1.71 ± 0.17a
3.22 ± 0.24
1.47 ± 0.18
2.83 ± 0.45
23.72 ± 1.20

a

2.54 ± 0.20
1.41 ± 0.18
3.12 ± 0.49
23.29 ± 1.39

F
15.43

P
<0.01

Denominator degrees of freedom were calculated with the Kenward-Roger method (SAS 9.2,
2002-2008).
b
Different letters indicate significant differences within row according to LSMeans (α=0.05).
c
Abundance determined from visual abundance per plot.
d
Abundance determined from sticky trap abundance per plot.
e
Abundance determined using model with marketable cabbage yield as a covariate.
f
Abundance determined using model with total herbivore abundance as a covariate.
g
Abundance determined using model with total predator abundance as a covariate.
h
Abundance determined using model with imported cabbageworm abundance as a covariate.
i
Abundance determined using model with Cotesia rubecula abundance as a covariate.

42

A)
Oat kill date:
Early
Intermediate
Late
Early
Intermediate
Late

Herbicide
Intensity:
Low
Low
Low
High
High
High

Early
Intermediate
Early
Intermediate
Late
Late

Low
Low
High
High
Low
High

2010

2011

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Maximum Distance Between Clusters
B)
Oat kill date:
Early
Intermediate
Late
Intermediate
Late
Late

Herbicide
Intensity:
Low
Low
High
High
Low
High

Early
Intermediate
Early
Intermediate
Late
Late

Low
High
High
Low
High
Low

2010

2011

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Maximum Distance Between Clusters
Figure 2.1. Habitat complexity dendograms based on average linkage distance for the herbivore
community (A) and natural enemy community (B) in 2010 and 2011.
43

Effect size (d± 95% C.I.)

0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

Herbivore feeding guild

2010

2010

2010

2010

2011

2011

2011

2011

Herbivore
Parasitoid
Predator

0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

Effect size (d± 95% C.I.)

Feeding functional group

Chewing

Sap Feeding

Diet breadth

Generalist
Specialist
Oligophagous

Hunting mode

Active

Sit and Wait

Figure 2.2. Effect of habitat complexity on insect abundance classified according to functional feeding groups, feeding guilds, diet
breadth, and hunting modes. Error bars indicate 95% bootstrap confidence intervals. Graphs in the top row are data from 2010, and in
the bottom row are from 2011.

44

Discussion
This study examined the impact of vegetational and structural complexity, created by the
presence of weeds and oat mulch in cabbage fields, on arthropod communities. The effects of
habitat complexity between cabbage rows on natural enemy and pest community structure,
individual insect species abundance, and insect guilds based on functional traits were
investigated.

Insect community structure. Changes in habitat complexity influenced community structure of
both herbivore and natural enemies in cabbage. In 2010, herbivore communities were strongly
influenced by herbicide intensity (Fig. 2.1A), which affected the abundance of weeds (Table 2.2).
In 2011, when weed pressure was lower, herbivore communities were influenced primarily by
oat kill date (Fig. 2.1A), which determined the amount of oat mulch between cabbage rows
(Table 2.2). This separation of herbivore communities along low and high complexity treatments
regardless of how complexity was achieved indicates that multiple methods of habitat
management can be used to shape herbivore community structure in this system. Natural enemy
communities in both years were more influenced by oat kill date than by herbicide intensity (Fig.
2.1B), which may be the result of natural enemy communities dominated by species that are
more responsive to indirect resources such as refuge. In general, these results support the idea
that most herbivores (Risch et al. 1983) and natural enemies (Langellotto and Denno 2004) are
responsive to changes in habitat complexity in agroecosystems.

Insect abundance. Diversified agroecosystems are expected to harbor fewer pests (Risch et al.
1983), but there can be significant deviations from this based on pest species and year. Contrary

45

to past studies in crucifer crops and onions, several herbivore species and natural enemies did not
respond to changes in habitat complexity, including the diamondback moth, multicolored Asian
lady beetle, and the parasitoid, Diadegma insulare (Zhao et al. 1992, Harvey and Eubanks 2004,
Alhmedi et al. 2009). These insects may be unresponsive to the types of structural or
vegetational complexity in this study or to changes at this spatial scale.

Impact of oat kill date on insect abundance. Crucifer flea beetles were positively
impacted by the mulch in between rows (Table 2.4), supporting observations by Schellhorn and
Sork (1997) in other cole crops in which flea beetles preferred complex and diverse habitats. On
the other hand, onion thrips responded negatively to mulching contrary to past observations in
mulched onions (Larentzaki et al. 2008).
In 2011, predators such as the pink lady beetle, fourteen spotted lady beetles, lady beetle
larvae, and predatory thrips responded positively to the presence of oat mulch in between rows
(Table 2.6). Although several of these species have responded similarly to mulching in other
crops, such as potato and sweet corn (Brust 1994, Lundgren et al. 2009), my results for predatory
thrips and lady beetles, are contrary to results from mulched versus non-mulched onion plots
(Larentzaki et al. 2008) and a meta-analysis conducted by Langellotto and Denno (2004).
The positive responses of arthropod abundance to greater mulch in the row-middle may
be due to provisioning of shelter, resources, or alternative prey in between rows (Root 1973,
Landis et al. 2000, Sunderland and Samu 2000).
In addition to the generalist predators in the cabbage fields, a specialized natural enemy
(Cotesia rubecula) was positively impacted by the later oat kill date, with greater abundance in
2011 (Table 2.6). Habitat complexity created by non-crop flowering plants can increase Cotesia

46

rubecula parasitism and abundance (Lee and Heimpel 2005), however the effects of non-living
habitat are unknown. Although Cotesia rubecula wasps rely on chemical cues from host plants
and visual cues to find its host (e.g. van Poecke et al. 2003), the greater oat mulch in row middles
did not seem to interfere with these cues. The superfamily Chalcidoidea consisting of
hyperparasitoids of D. insulare and C. rubecula, were also more abundant in the plots with later
oat kill dates in 2010 and 2011, and lower herbicide intensity in 2010 (Table 2.6, 2.7). Mulching
appears to positively influence the third and fourth trophic levels. Cotesia rubecula and
Chalcidoidea hyperparasitoids may find shelter and attractive microclimates in mulched plots.

Impact of herbicide intensity on insect abundance. Increasing vegetational complexity
had positive effects on the specialist imported cabbageworm in past studies (Latheef and Ortiz
1983, Maguire 1984). However, for several key herbivore species, the findings of a negative
relationship between imported cabbageworm, cabbage aphid, and cabbage looper abundance and
herbicide intensity (Table 2.5), supports previous results from broccoli, cauliflower, and collards
with vegetational complexity (Altieri et al. 1985, Altieri and Schmidt 1987, Garcia and Altieri
1992, Schellhorn and Sork 1997, Norris and Kogan 2000, Hooks and Johnson 2002). In these
cases, complex row middles due to reduced herbicide intensity may reduce the encounter rate
with the host plant and eventually reduce the amount of time spent on cabbage (Finch and Collier
2000). If weed species are preferred over an adjacent crop, the weeds may act as a barrier or a
trap crop (Andow 1988).
Predators were generally positively impacted by vegetational complexity in between
cabbage rows. Predator species, including the spined soldier bug, minute pirate bug, pink lady

47

beetle, and fourteen spotted lady beetle were more abundant in plots with greater weed pressure
in 2010 (Table 2.7).
Although the exact mechanisms for these results, including positive responses to habitat
complexity, are yet unknown, they support suggestions that natural enemies may find shelter,
additional resources, or alternative prey within these complex habitats (Root 1973, Landis et al.
2000, Sunderland and Samu 2000). The enemies hypothesis (Elton 1958) suggests that increased
habitat complexity will increase attack rates of herbivores by natural enemies, however in other
studies in cabbage, reduced predation and parasitism has been observed in intercropped plots and
higher biocontrol in monocultures (Bjorkman et al. 2010, Adati et al. 2011). Complex habitats
with non-crop plants such as weeds can host a greater diversity of herbivores, supporting a
greater diversity of natural enemies within an agricultural field (Andow 1988). For instance, the
weeds and oat mulch in between cabbage rows may be habitat for alternative prey for natural
enemies.

Significant covariates for herbivores. In addition to the direct impacts of habitat
complexity on arthropod abundance (Figure 1.3), impacts on marketable yield, predator
abundance, and abundance of prey or hosts can indirectly influence arthropod abundance across
trophic levels (Figure 1.4). Cabbage yield was reduced by later oat kill dates and reduced weed
management. The amount of marketable cabbage yield in turn positively influenced cabbage
aphids and negatively influenced crucifer flea beetles and imported cabbageworm abundance
during this study. Changes in cabbage head size and cabbage quality appear to influence habitat
choice for these herbivores. Total predator abundance or the third trophic level influenced
abundance of imported cabbageworm, diamondback moth, cabbage loopers, and onion thrips

48

suggesting predation or changes in predator efficiency may be playing a role in driving prey
density in the field. For instance, abundance of the imported cabbageworm was influenced by
weed management. Due to the high mobility and egg dispersal of this pest, Banks (1998) has
suggested any observed differences in imported cabbageworm abundance across habitats are due
to differences in predation. This suggests predation of the imported cabbageworm is lower in the
simple habitats with low weed pressures. Since total predator abundance was positively related to
imported cabbageworm abundance, total predator abundance does not appear to be a reliable
indicator of predation. Perhaps a few key predators are responsible for imported cabbageworm
predation.

Significant covariates for natural enemies. Cabbage yield consistently negatively
impacted pink lady beetle abundance. Larger cabbage plants may increase hunting time and
decrease the lady beetle’s hunting efficiency. The quality of odor signals could also have been
impacted by cabbage yield or quality, although this was not examined during this study. Total
herbivore abundance was only included in the model for the pink lady beetle, suggesting that
most generalist predators are not choosing habitats based primarily on prey availability in the
field. On the other hand, abundance of C. rubecula and its hyperparasitoids were influenced by
their respective host’s abundance. It is important to note that these specialists appear to follow
their hosts around the field and may be relying on odor and visual cues from its hosts, although
host searching for C. rubecula and the Chalcidoidea hyperparasitoids were not impeded by
mulching or weed pressures. Both abundance of generalist predators (e.g. lady beetles and
predatory thrips) and specialist parasitoids (e.g. C. rubecula and D. insulare) were impacted by

49

total predator abundance suggesting that intraguild predation and interactions within the natural
enemy guild could be impacting natural enemy populations in the cabbage field.

Insect functional groups. Research in other agricultural systems (e.g. wheat, canola, cover
crops, and corn) and in natural habitats have often focused on the relationship between habitat
complexity and individual arthropod species or a single group of arthropods, such as ground
beetles (Carmona and Landis 1999, de la Pena et al. 2003, Harvey et al. 2008, Hummel et al.
2012). The use of functional traits is suggested to provide a view of communities, while
providing information beyond the species level (McGill et al. 2006). Categorical classification
can also provide an understanding of how treatments influence insect groups, particularly when
every insect cannot be classified to species (Buschke and Seaman 2011). Studies that examined
the influence of vegetational complexity on natural enemy communities in agroecosystems or
natural habitats, have determined that vegetational complexity affects insect community structure
across multiple trophic levels or functional groups (Tylianakis et al. 2007, Haddad et al. 2011).
In this study, high complexity plots positively influenced predators in 2011, according to a
significant positive effect size (Figure 2.2). This appears to support the positive responses of
predator abundance to the habitat complexity treatments and previous findings of greater
predator abundance in cabbage plots with mulch (Xu et al. 2011) or weed communities
(Schellhorn and Sork 1997). However, we found that the impact of habitat complexity treatments
did not significantly differ among insect life stages or their location in the field, thus these traits
have low priority when ranking them for responsiveness to habitat complexity manipulation
(McGill et al. 2006). Higher priority functional traits included feeding guild, diet breadth, and
hunting mode. In 2011, the high habitat complexity treatments negatively affected chewing

50

insects, as well as oligophagous and specialized feeders, suggesting that generalist herbivores
including sap feeders benefited the most from habitat complexity (Figure 2.2). This trend has
also been found in natural habitats, including grasslands, with greater specialist herbivore
richness within heavily grazed grasslands with low structural complexity and greater generalist
herbivore abundance in heterogeneous environments at the landscape level (Batary et al. 2007).
In 2011, active hunters were positively influenced by increased habitat complexity and sit and
wait hunters were negatively impacted (Figure 2.2). This does not appear to support past
observations, in which weeds and mulch increased the presence of both ground dwelling and
web building spiders (Altieri et al. 1985, Riechert and Bishop 1990, Hooks and Johnson 2004,
Langellotto and Denno 2004). As supported by the large number of web building spiders located
on the oats in my field, and the species’ known preference for structural complexity (Rypstra et
al. 1999), this particular negative response of sit and wait hunters may be the result of my
restricted sampling method on cabbage, rather than true reductions in hunter abundance.
Behavioral traits, including functional feeding group, feeding guild, diet breadth, and hunting
mode appear to be responsive traits to habitat management and should be considered in future
studies.

Summary. As indicated by significant differences in weed biomass, oat height, and oat biomass
across treatments, cabbage fields in 2010 and 2011 contained habitats with varying degrees of
complexity between rows. This study suggests that greater habitat complexity between cabbage
rows can enhance predator populations. However, this complexity reduced cabbage yield,
increased populations of a few pests, and enhanced hyperparasitoid abundance, which may
reduce the efficiency of the natural enemies in providing effective biological control within the

51

field. Possible indirect effects of mulching and weed management were also observed, with
marketable yield, herbivore abundance, and natural enemy abundance influencing the abundance
of individual arthropod species. In particular, cabbage yield and quality, host or prey abundance,
and predation appear to play a role in individual species abundance. Cabbage biomass and yields,
mid-season and at-harvest respectively, were reduced due to interactions between cabbage,
weeds, and oats in 2010. Other studies have also observed cabbage yield reductions in killed oat
plots (Bottenberg et al. 1997, Mochizuki et al. 2008). Further studies are needed to determine
management strategies that improve arthropod management while minimizing weed and cover
crop interference with the crop. It would also be helpful to determine the exact mechanisms
responsible for the observed arthropod responses across mulching and weed pressure gradients.

52

CHAPTER 3.
The impact of mulch species and management on
natural enemies and biological control in cabbage

Introduction. Cabbage production in the Midwestern United States is threatened by a diverse
community of pests, including the imported cabbageworm (Pieris rapae, Lepidoptera: Pieridae),
diamondback moth (Plutella xylostella, Lepidoptera: Plutellidae), cabbage looper (Trichoplusia
ni, Lepidoptera: Noctuidae), flea beetles (Coleoptera: Chrysomelidae), and aphids (Hemiptera:
Aphididae). In cabbage, high levels of predation and parasitism, up to 80% and 50% respectively,
have been observed for the imported cabbageworm (Schmaedisk and Shelton 1999, WoldBurkness et al. 2005). This suggests natural enemies can provide substantial biological control
within a cabbage field. Conventional pest management practices, including monocultures and
broad-spectrum insecticides can reduce natural enemy populations and reduce the services they
provide in vegetables (Bommarco et al. 2011). In contrast, habitat management techniques such
as cover cropping have been shown to enhance biological control in vegetables and field crops
through the conservation of these natural enemies (Lundgren et al. 2011, Schmidt et al. 2004,
Schmidt et al. 2007).
Cover crops are non-harvested plants that provide habitat and resources for natural
enemies (Carmona and Landis 1999). When allowed to remain in the cabbage field as living
mulch, these plants have been associated with reductions in insect pest abundance (Andow et al.
1986). Past studies have examined other benefits of living mulches and cover crops incorporated
in crop rotations such as nutrient uptake (Bath et al. 2006), reductions in nutrient leaching and
runoff (Martinez and Guiraud 1990), greater soil health (Teasdale 1998), and weed reduction

53

(Ilnicki and Enache 1992). Competition between the cover crop and the cash crop has also led to
reductions in cabbage yield (Brainard et al. 2004) and limited the adoption of this conservation
method.
Killing the cover crop prior to cabbage transplantation may provide an alternative to
living mulch. Limiting tillage to planting rows allows killed cover crop mulch to remain on the
soil surface. The benefits of cover crop mulch between cabbage rows include reduced soil
temperatures and conservation of soil moisture (Schonbeck et al. 1993). Mulching also reduces
weed infestations (Bottenberg et al. 1997, Brandsaeter et al. 1998, Franczuk et al. 2010,
Masiunas et al. 1997, Zaniewicz-Bajkowska et al. 2009) and can provide weed suppression
comparable to herbicide applications (Dillard et al. 2004, Mangan et al. 1995). Effects of cover
crop mulching on cabbage yields has been mixed. While several studies have observed decreased
cabbage weight (Bottenberg et al. 1997, Brandsaeter et al. 1998, Trdan et al. 2008), mulching has
maintained (Haramoto and Brainard 2012) or enhanced cabbage yields in other cases (Franczuk
et al. 2010, Riley et al. 2003).
Killed cover crop mulches and other plant residues can reduce diamondback moth,
imported cabbageworm, cabbage looper, and aphid abundance in cabbage (Bottenberg et al.
1997, Mangan et al. 1995, Xu et al. 2011). Fewer studies have examined the impact of these
management strategies on the natural enemy community. Xu et al. (2011) found greater natural
enemy densities in cabbage plots with plant residues present on the soil surface compared to bare
soils, suggesting mulching can impact conservation biological control within the field. However,
the impact of different mulch species and the management of these mulches on biological control
services such as predation and parasitism have not been examined in cabbage.

54

When choosing cover crops to conserve natural enemies, two groups, legumes and cereal
crops, have been more commonly adopted by growers in Michigan and show success in
enhancing organic matter, pest management, yield and quality (Snapp et al. 2005). These distinct
groups of cover crops have physical or structural differences, which may have diverse impacts
on natural enemies through the creation of different types of habitat complexity.
Structural complexity is non-living structure within a habitat that can be manipulated (e.g.
surface area, density, height) through the use of different species of cover crop mulches. For
instance, legumes are broadleaf plants with greater leaf area than cereal plants. Many legume
cover crops (e.g. clovers) grow along the soil surface, providing habitat for arthropods close to
the ground. In contrast, many cereal cover crops create greater vertical habitat when allowed to
remain standing. Past studies have found that greater structural complexity within a habitat and
within a plant enhances natural enemy populations through provisioning of alternative resources,
increasing predation efficiency, and providing shelter from intraguild predators (Langellotto and
Denno 2004).
Structural complexity can also be manipulated through the management of these cover
crops. For example, rolling and crimping reduces cover crop height and the vertical complexity
of the habitat between cabbage rows, while creating dense mulch habitat along the soil surface.
Studies on management strategies for cover crop mulch have focused on impacts on the first
trophic level and abiotic factors, excluding insect pests and natural enemies. For instance, rolling
mulch has shown to reduce soil temperatures and has a mixed effect on cabbage yields varying
from year to year, while the impact on natural enemies is unknown (Mochizuki et al. 2008). In
one case, Laub and Luna (1992) found that reducing cover crop mulch height in corn with
mowing (allowing mulch to remain standing) led to greater natural enemy abundance (e.g.

55

ground dwelling predators) compared to plots with cover crop mulch lying on the soil surface
(Laub and Luna 1992). However, mulching management strategies that manipulate habitat height
between cabbage rows has never been examined for impacts on biological control.
Monitoring of pest and natural enemy populations in the field provides information on
insect diversity and abundance across habitat managements. It does not necessarily explain the
mechanism for population reductions and cannot be linked with predation or parasitism rates
(Furlong and Zalucki 2010). Sentinel prey studies with natural enemy exclusion cages
(Costamagna et al. 2008) can provide predation and parasitism rates provided by guilds of
natural enemies. Sentinel prey studies also provide information on the biological control of a
specific life stage of the pest species. However, individual natural enemy species providing
significant reductions in the pest population cannot be identified with this method.
The goals of this study were to examine the impacts of cover crop species (e.g. legume
and cereal) and management (crimped or no crimping) on 1) total arthropod abundance 2)
abundance of individual arthropod species and 3) biological control for one of the major cabbage
pests in Michigan, the imported cabbageworm. Complex habitats with standing cereal mulch
were predicted to positively impact the natural enemy community and individual species.
Enhanced biological control was also predicted in these parts of the field.

Materials and Methods
Experimental Design. A field study was established in fall 2011 at the Michigan State
University’s Horticulture Farm in East Lansing, Michigan. The experiment was a randomized
complete block design with six treatments, consisting of combinations of two main effects:
mulch species (none, vetch, or rye) and management (none or crimping between rows).

56

Treatments were replicated four times. Plots (9.14 m x 9.14 m) consisted of 12 rows of cabbage
with 76 cm row spacing. Two buffer rows were planted along the edges of the field.

Field Operations. On October 4, 2011 cover crop seeds were broadcast by hand. Winter
rye (Secale cereale; Dirt Works, New Haven, VT) was broadcast in rye plots at a rate of 183.10
Kg/ha. Hairy vetch (Vicia villosa; Johnny’s Selected Seeds, Albion, ME) was broadcast in vetch
plots at a rate of 61 Kg/ha. On April 27, 2012 Select Max (2.34 L/ha) was applied with a nonionic surfactant (Dispatch; 1.17 L/ha) to kill the winter rye cover crop. To kill the hairy vetch
cover crop, Roundup (3.51 L/ha) was applied on May 11 followed by an application of Stinger
(0.58 L/ha), applied on May 22 to effectively kill the vetch cover crop. Due to the dense biomass
in vetch plots, vetch plots were mowed using a push mower (DR® Trimmer; Vergennes, VT)
reducing vetch biomass and allowing passage of the strip-tiller. On May 29, a Hiniker® Model
6000 two-row strip-tiller (equipped with cutting-coulter, shank-point assembly, and rolling
basket) was used to create 25 cm x 25 cm deep strips at 76 cm between-strip spacing. According
to soil tests and nutrient management recommendations for cabbage, 345 kg/ha of 19-19-19
(NPK) fertilizer was applied in planting rows at this time (Warncke et al. 2004). Nitrogen was
not sidedressed later in the season.
Cover crops were killed prior to strip-tillage. Strip tillage allowed cover crop residue
undisturbed in approximately 51 cm between planting rows. After tillage, plots assigned to
crimping treatments were crimped by hand using a plywood rectangle (76 x 25 cm) equipped
with rope handles. This hand-made crimper effectively knocked over standing mulch in crimped
plots.

57

Cabbage transplants (var. “Blue Dynasty”; Siegers Seed Company, Holland, MI) were
grown for 4 wk in 98 cell plug trays and transported to the Horticulture Farm to harden off for 15
d prior to transplanting in the field. Cabbage was transplanted by hand beginning on May 29 into
strip-tilled zones (76 cm between rows) with 36 cm between transplants within the row. No
fungicides were used. Herbicides included GoalTender 4SC (0.15 L/ha) and Select Max (0.73
L/ha) applied on 10 July across the entire field to manage weeds. By July 8, vetch mulch had
decomposed and was no longer visible on the soil surface. To further manage weeds at this time,
row middles in vetch plots and plots without mulch were cultivated using a two shank cultivator
pulled behind a tractor (John Deere® model 5325; John Deere Company, Moline, IL). An
herbicide, GoalTender 4SC (0.44 L/ha), was applied throughout the field and additional handweeding was performed in all plots at this time to reduce weed pressure further.

Mulch, weed, and cabbage yield measurements. Mulch height and biomass were
measured three times during the growing season on June 18, August 8, and September 4. Mulch
2

height was measured in two 0.25 m quadrats per plot. Within these randomly placed quadrats,
ten height measurements were recorded and mulch biomass was collected. Prior to weighing, all
above ground mulch biomass was dried in the oven at 60°C for 7d.
A weed census was also conducted 40d prior to handweeding and 8d and 36d after
2

handweeding. Within the 0.25 m quadrats, weeds were counted and identified to species. Based
on census data, weed abundance and diversity (Shannon-Wiener Index) were calculated.
Maximum height for each weed species was also recorded. The most common weed species
included: nutsedge (Cyperus esculentus L.), common lambsquarters (Chenopodium album L.),

58

smartweed (Polygonum pensylvanicum L.), common ragweed (Ambrosia artemisiifolia L.),
redroot pigweed (Amaranthus retroflexus L.), and wild radish (Raphanus raphinistrum L.).
Cabbage was harvested beginning on September 5. In each plot, cabbage heads were
harvested from the middle 4.57 m of the central 8 rows. The number and weight of marketable
and non-marketable heads was also assessed. Cabbage heads greater than 11.4 cm in diameter
were considered marketable.

Mulch, weed, and cabbage yield statistical analysis. A mixed model (PROC Mixed,
SAS 9.2, 2002-2008) was used to determine the effects of mulch species and management on
mulch height, mulch biomass, and cabbage yield. The variable, mulch height was averaged per
plot. Mulch biomass and cabbage yield (e.g. biomass and number of marketable and nonmarketable heads) were summed per plot. Mulch height, mulch biomass, and cabbage yield were
square root transformed (sqrt(x + 0.5)) to meet assumptions of normality and equal variance. An
analysis of covariance was performed with a model including mulch species, mulch management,
and their interaction as fixed factors and block as a random factor. Weed species height and
diversity were included as covariates to determine the impact of these factors on insect
abundance and cabbage yield. A least-square means test with Tukey-Kramer adjustment was
performed if main effects or their interactions were significant (α=0.05).

Arthropod sampling. To determine the effect of mulch species and management on
arthropod abundance, Insects were sampled using three sampling methods. First, sessile insects
were sampled visually on cabbage plants. Insects were identified and counted on 10 randomly
chosen cabbage plants per plot for 11wk during the growing season. Plants were chosen from

59

central rows within the plot, eliminating edge effects. All above ground plant tissue was
examined during visual sampling.
Highly mobile insects were sampled with yellow sticky traps on metal stakes (7.5 × 12.3
cm, Great Lakes IPM, Vestaburg, MI) for 10 wk during the growing season. Sticky traps placed
0.4 m from ground level and changed every 7 d.
Ground dwelling arthropods were sampled using pitfall traps 7 times throughout the
growing season. Four plastic cups (354.88 ml; Solo Cups, Lake Forest, Illinois) were placed
systematically in the central rows of each plot. Traps were submerged in the ground until the
brim of the cup was flush with the surface of the soil. These traps remained uncovered for 48 hr
before arthropods were removed and identified. All arthropods were identified to species or the
lowest possible taxonomic classification (Marshall 2006). Voucher specimens for the imported
cabbageworm and major natural enemies were deposited in the A.J. Cook Arthropod Research
Collection at Michigan State University (East Lansing, MI).

Arthropod communities. Abundance data from visual, sticky trap, and pitfall sampling
was transformed using log (x + 0.01) to meet assumptions of normality and analyzed using a
covariate analysis. Total mobile predators on sticky traps, total ground dwelling predators in
pitfall traps, and total parasitoids on sticky traps were examined. Mulch species, mulch
management, and the interaction between mulch species and mulch management were fixed
factors, block as a random factor, treatment nested within block as the repeated subject (PROC
Mixed). Covariates such as weed height, weed diversity, and marketable cabbage yield were
included using data at the closet sampling date. A least-square means test with Tukey-Kramer
adjustment was performed if main effects were significant (α=0.05).

60

Arthropod abundance. Abundance data from insect sampling was log transformed
using log (x + 0.01) to meet normality assumptions and analyzed with a repeated measures
analysis. Impact of treatments on individual herbivore species abundance were examined using
on visual sampling and natural enemy data was taken from sticky trap and pitfall trap sampling
data. Models included mulch species, mulch management, and the interaction between mulch
species and mulch management as fixed factors, date as a repeated factor, block as a random
factor, and treatment nested within block as the repeated subject (PROC Mixed). Covariates
included weed height, weed diversity, marketable cabbage yield, total herbivore abundance, total
predator abundance, total parasitoid abundance, and host abundance. For instance, diamondback
moth abundance was included as a covariate for analyzing Diadegma insulare abundance and
imported cabbageworm abundance was included as a covariate for Cotesia rubecula abundance.
If main effects were significant (α=0.05), we performed least-square means tests with TukeyKramer adjustment.

Natural enemy exclusion cages. To determine treatment effects on biological control
services provided by different natural enemy guilds (e.g. small predator, large predator,
parasitoid), Exclusion cages were placed in the field with sentinel prey. Four exclusion cages
were placed in each plot. Each cage consisted of an 8 wk old cabbage plant in a 15.24 x 13.97
cm (diameter x height) pot with Sure-Mix Perlite (Michigan Grower Products, Galesburg, MI).
Two 0.91 m wires bent and submerged into the potting soil created a wire framework
surrounding the cabbage plant. Four transplanted cabbage plants were randomly selected and
replaced by a caged cabbage plant. Each cage was submerged in the soil and leaf pieces with
four first instar imported cabbageworm larvae were transferred to the caged plant. Plants were

61

infested with imported cabbageworm from a greenhouse colony at Michigan State University.
Larvae were reared on brussel sprouts (Brassica oleracea; Johnny Select Seeds, Albion, ME)
with a 16:8 h, L:D photoperiod.
The first cage, excluded all arthropods with a fine white no-see-um netting (0.002 mm

2

holes, Skeeta Company, Bradenton, FL) closed with a clip at the top of the frame and closed with
a rubber band near the bottom of the pot. Any mortality of prey in this cage could be associated
with natural mortality due to environmental conditions or handling. The second cage, a
mosquito-netting cage, had a white mosquito netting (Skeeta Company, Bradenton, FL) cover
instead. This exclusion cage had 2 mm openings, allowing parasitoids and small predators access
to the sentinel prey but exclude larger predators, including predatory stink bugs and lady beetles,
were excluded. The third cage type, a sham cage, was identical to the mosquito-netting cage.
However, the sham cage was not closed with a rubber band at the bottom. The netting was raised
1 cm above the soil surface allowing ground predators access to the sentinel prey. The netting
also allowed larger predators access to prey through 10 rectangular openings (3 x 20 cm) in two
rings around the netting. The sham cage accounts for any effects of the presence of netting on
insect movement. The final cage, an open control cage lacked any netting allowing movement of
any size natural enemy into the cage.
Cages were placed in the field for a 36 h period before retrieval. After 36 h, predators
were removed from cages and plants were re-covered with a fine mesh. Any remaining imported
cabbageworm larvae were reared on cabbage plants until pupating into an adult butterfly or a
parasitoid pupa or adult was observed. Predation and parasitism rates were calculated based on
larvae killed or parasitized. The exclusion cage experiment was repeated three times during the
growing season starting on June 28, July 30, and August 23.

62

Mortality and parasitism rates were transformed using sqrt(x + 0.5) to meet assumptions
for normality and analyzed using an analysis of variance with block as a random factor,
treatment nested within block as the subject, and mulch species, mulch management, cage type,
and the interactions of these variables as fixed factors (PROC Mixed). A least-square means test
with Tukey-Kramer adjustment was performed if main effects were significant (α=0.05).

Results
Mulch, weed, and cabbage yield measurements. Mulch height was significantly impacted by
mulch species, mulch management, and the interaction between these factors throughout the
cabbage growing season (Table 3.1). At an early, intermediate, and late date in the growing
season, mulch height in rye plots were significantly higher than vetch (t > 7.40, df = 15, P <
0.02) or bare soil plots (t > 7.40, df = 15, P < 0.01). Mulch height in vetch plots was not
significantly different from bare soil plots throughout the growing season (t = 1.88, df = 15, P >
0.08). Throughout the summer, plots without crimping contained taller mulch than crimped plots
(t > 2.54, df = 15, P < 0.02). However, only rye plots contained lower mulch in crimped plots
compared to non-crimped plots (t > 4.40, df = 15, P < 0.01). Rye mulch was approximately three,
11, and four times taller in non-crimped compared to crimped plots during the early, intermediate,
and late portions of the season respectively.
Mulch management (crimping) and the interaction between mulch species and
management did not affect mulch biomass (Table 3.1). However, mulch biomass differed across
mulch species treatments (Table 3.1), with the greatest amount of biomass in rye plots
throughout the growing season. During the early, intermediate, and late parts of the season,
approximately two, 21, and 18 times greater biomass was recorded in rye plots than vetch plots.

63

Mulch management did not influence weed height, weed abundance, or weed species
richness (F < 1.18, df = 1,18, P > 0.29). However, weed height was significantly reduced in plots
with rye mulch (Table 3.2), with weed height in rye plots 48% and 40% lower than bare soil and
vetch plots respectively (t = 2.75, df = 18, P < 0.03). Weed abundance and species richness were
2 times and 50% greater in bare soil plots respectively compared to rye mulch plots (t >3.19, df =
15, P < 0.02).
Cabbage yields at harvest were divided into marketable and non-marketable yields (Table
3.3, Supp Table S4). The covariates, weed height and diversity, during the early, intermediate,
and late sampling date were not significant and therefore were excluded from the cabbage yield
models. Mulch management did not significantly affect cabbage yields (t < 0.28, df = 15, P >
0.96), but marketable yield was two times greater in bare soil and vetch plots compared to plots
with rye mulch (t > 2.57, df = 15, P < 0.05). In contrast, non-marketable yields were greatest in
rye plots with 1.5 times more cabbage biomass compared to bare soil or vetch plots (t > 3.11, df
= 15, P < 0.02).
The number of marketable and non-marketable cabbage heads did not differ across mulch
management treatments. However, the number of cabbage heads was significantly different
across mulch species (Supp Table S4). Bare soil plots contained the greatest number of
marketable cabbage heads, with two times more heads than rye plots (t = 3.31, df = 15, P = 0.01).
Alternatively, rye plots contained the greatest number of non-marketable cabbage heads, with 1.5
times more cabbage heads than bare soil plots (t = 3.13, df = 15, P = 0.02). The number of
marketable or non-marketable cabbage heads in vetch plots was not significantly different from
the other treatments (t < 2.39, df = 15, P > 0.07).

64

Arthropod communities. Weed diversity affected total mobile predator abundance on July 1,
July 23, and August 11 (F >5.41; df = 1,24; P < 0.03). Neither weed diversity nor weed height
were significant covariates for total ground predator or total parasitoid abundance and were
excluded from both models.
Trends were not consistant for total mobile predators, ground predators, or parasitoids
throughout the growing season, suggesting a limited impact of habitat complexity on total natural
enemy abundance. For instance, on July 1, mobile predators were 51% more abundant in bare
soil plots compared to rye plots (t >3.16, df = 24, P < 0.01). However, on June 18 and August 11,
predators were 86 and 33% more abundant in rye plots compared to bare soil (t > 3.16, df = 24, P
< 0.01). Mulch management was significant on June 26, July 16, and August 16 (F > 7.62; df =
1,20; P < 0.02) with 27 and 49% more predators in non-crimped plots in June 26 and July 16 (t
>2.76, df = 20, P < 0.02), while crimped plots contained 67% more predators on August 16 (t =
2.81, df = 19, P = 0.01).
Total ground predator abundance significantly differed among treatments at the end of
the season on the August 30 sampling date (Figure 3.2). Total ground beetle abundance in rye
plots without crimping was more than 8 times greater than vetch plots without crimping (t = 4.45,
df = 15, P < 0.01) and crimped bare soil plots (t >3.44, df = 15, P =0.04).
Trends in total parasitoid abundance were not consistant throughout the season. For
instance, plots with rye or vetch mulch contained 62 and 51% more parasitoids than bare soil
plots (t >2.99, df = 15, P < 0.02). On July 16 total parasitoid abundance differed among mulch
management (F = 11.58; df = 1,12; P < 0.01) with 26% fewer parasitoids in crimped compared
to non crimped plots (t = -3.40, df = 12, P < 0.01). Although bare soil crimped plots contained
remarkably high parasitoid abundance compared to other treatments on July 1 and 10, parasitoid

65

abundance was not significantly higher in these plots across mulch species (F < 1.26; df = 2,15;
P > 0.31), mulch management (F < 2.78; df = 1,15; P > 0.12), or the interaction between species
and management (F < 2.70; df = 2,15; P > 0.10). A group of gregarious pupal parasitoids
(Chalcidoidea) accounted for 72-99% of the parasitoids on these two sampling dates. However,
abundance of this species ranged greatly (38-385 individuals per plot). In particular, a single
experimental plot in the corner of the field contained 385 and 343 individuals of this species on
July 1 and 10 accounting for the spike in parasitoid abundance.

Arthropod abundance. Since interactions between main effects were not significant, only main
effects are presented here. Mulch management did not affect the abundance of pests such as the
imported cabbageworm, diamondback moth, cabbage looper, cabbage aphid, crucifer flea beetle,
and onion thrips (Table 3.5). One pest, the striped flea beetle, was 13% more abundant in noncrimped plots compared to crimped plots (t = 2.63, df = 21, P = 0.02). Abundance of a single
natural enemy species, the fourteen spotted lady beetle, also differed across mulch management
(Table 3.6). This predator was 2.5 times more abundant in crimped compared to non-crimped
plots (t = 2.57, df = 20, P = 0.03).
While most pests did not respond to mulch species, the imported cabbageworm (Table
3.7) were 20% more abundant in rye compared to hairy vetch plots (t = 5.10, df = 21, P < 0.01)
and 19% more abundant in rye compared to bare soil plots (t = 4.91, df = 23, P < 0.01).
Mulch species affected the abundance of several natural enemy species (Table 3.8). Three
natural enemies were negatively impacted by the presence of rye mulch, including the pink lady
beetle, and Poecilus sp. ground beetles other than Poecilus chalcites. The pink lady beetle was
16% and 33% more abundant in bare soil (t = 2.94, df = 28, P < 0.01) and hairy vetch plots (t =

66

2.19, df = 27, P = 0.04) respectively, compared to plots with rye mulch. Ground beetles in the
genus Poecilus, excluding the common Poecilus chalcites species, were seven and three times
more abundant in bare soil (t = 5.30, df = 27, P < 0.01) and hairy vetch plots (t = 3.26, df = 23, P
< 0.01) respectively, compared to rye plots.
Five natural enemies were positively impacted by the presence of rye mulch, including
four generalist predators (Harpalus pennsylvanicus, Pterostichus permundus, and spiders in the
family Linyphiidae) and a specialist parasitoid, Cotesia rubecula. The ground beetle Harpalus
pennsylvanicus was nearly three times and five times more abundant in rye plots compared to
hairy vetch (t = 4.42, df = 18, P < 0.01) and bare soil plots (t = 7.15, df = 22, P < 0.01)
respectively. The ground beetle Pterostichus permundus was nearly six times and 15 times more
abundant in rye plots compared to hairy vetch (t = 6.09, df = 21, P < 0.01) and bare soil plots (t =
5.66, df = 25, P < 0.01) respectively. A group of spiders in the family Linyphiidae commonly
referred as sheet weaver spiders were more than five times and two times more abundant in rye
plots compared to bare soil (t = 4.54, df = 22, P < 0.01) and hairy vetch plots (t = 3.18, df = 18, P
< 0.01) respectively. The specialist parasitoid, Cotesia rubecula, was approximately two times
more abundant in rye plots compared to plots with bare soil (t = 5.85, df = 26, P < 0.01) and 34%
more abundant in rye plots compared to vetch (t = 4.72, df = 24, P < 0.01).

Significant covariates for herbivores. Weed height was a significant covariate for the
imported cabbageworm, diamondback moth, cabbage looper, and striped flea beetle (F = 15.85,
P < 0.01) positively influencing herbivore abundance, although the imported cabbageworm was
negatively influenced by greater weed height. Weed diversity negatively impacted cabbage
looper and cabbage aphid abundace, and positively influenced crucifer flea beetle abundance (F

67

> 6.43, P < 0.01). Marketable cabbage yield positively impacted a single herbivore, the
diamondback moth (F = 8.24, df = 1,22, P < 0.01). Total predator abundance positively impacted
the imported cabbageworm and cabbage looper, and negatively influenced the diamondback
moth (F > 8.45, P < 0.01). Total parasitoid abundance positively impacted the imported
cabbageworm (F = 129.36, df = 1,257, P < 0.01).

Significant covariates for natural enemies. Weed height positively influenced the pink
lady beetle and negatively influenced C. rubecula (F > 34.68, P < 0.01). Weed diversity
positively impacted the spined soldier bug, minute pirate bug, and Lycosids, and negatively
impacted Cyclotrachelus sodalist and Pterostichus permundus (F > 6.43, P < 0.01). Marketable
cabbage yield significantly impacted Cyclotrachelus sodalist negatively and Anisodactylus
sanctaecrucis abundance positively (F > 7.55, P < 0.01). Total herbivore abundance was not a
significant covariate for any natural enemies, however total predator abundance significantly
impacted the ground dwelling arthropods: B. quadrimaculatum, Poecilus sp., Scarites quadriceps,
Anisodactylus sanctaecrucis, Harpalus pennsylvanicus, Amara sp., Cicindela sp., Staphylinidae,
Centipede, Lycosidae, and Phalangidae. As in 2010 and 2011, Cotesia rubecula and its
hyperparasitoids were positively impacted by abundance of their hosts (F > 3.84, P < 0.05).

Natural enemy exclusion cages.
Mortality. Mortality rates of the imported cabbageworm were not significantly different
across the interactions between cage types and mulch species (F < 1.49; df = 6,54; P > 0.20) or
cage types and mulch management (F < 0.63; df = 3,54; P > 0.60). This suggests larval mortality

68

due to different natural enemy guilds was unaffected by mulch species or management
treatments. Predation by small and large predators is suspected to be similar across treatments.
However, mortality rates were significantly different across cage types at each sampling
date (F > 18.51; df = 3,54; P < 0.01) (Figure 3.4). The no-see-um netting cages effectively
excluded all natural enemies with mortality rates between 1-5% during the season (Figure 3.4).
Predators and parasitoids were not observed inside the cages, suggesting loss of larvae were due
to natural mortality rather than actual predation. Throughout the season, mortality rates were
significantly lower in no-see-um cages compared to the other cage types (t > 5.41, df = 54, P <
0.01) with mortality rates ranging from 43-89%. Corpses were rarely recovered; this suggests
mortality may be due to movement of larvae off the plant in addition to consumption of the first
instar larvae. During the early, intermediate, and late part of the growing season, mortality rates
in the mosquito netting cage that excluded large predators, were not significantly different from
the sham or no-netting cages (t < 1.59, df = 54, P > 0.12) that allowed all natural enemies access
to the prey, suggesting the majority of larval predation is due to the small predator guild.

Parasitism. The interaction between cage type and mulch species (F < 2.13; df = 6,54; P
> 0.06) or cage type and mulch management treatments (F < 1.34; df = 3,54; P > 0.27) did not
affect the percent of parasitized imported cabbageworm. This suggests different sizes of
parasitoids responded similarly across treatments and mulch species or management did not
enhance parasitism by a particular guild of parasitoids.
The no-see-um netting cage effectively excluded parasitoids, with 0% parasitism
observed at each sampling date (Figure 3.5). During the early part of the season, parasitism rates
did not differ across cage type (F = 2.13; df = 3,54; P = 0.11). However, later in the growing

69

season, mortality rates were significantly different across cage types (F > 8.59; df = 3,54; P <
0.01). During the intermediate and late season sampling dates, parasitism rates in the mosquito,
sham, and no-netting cages (t > 3.24, df = 54, P < 0.01) which allowed parasitoids access to the
larvae, were significantly greater than the no-see-um cages which excluded the parasitoids.
Parasitism rates ranged from 18-25% later in the growing season.

70

Table 3.1. Mean ± SEM mulch height at an early, intermediate, and late date during the growing season.
Main effects
Treatment
Bare soil / no
crimping
Hairy vetch / no
crimping
Winter rye / no
crimping
Bare soil /
crimping
Hairy vetch /
crimping
Winter rye /
crimping

Early season

Mulch height (cm)
Intermediate
Season

a,b

Late season

0.00 ± 0.00c

0.00 ± 0.00c

0.00 ± 0.00b

1.53 ± 0.21c

0.38 ± 0.13bc

0.00 ± 0.00b

34.18 ± 8.18a

11.18 ± 2.03a

5.95 ± 1.81a

0.00 ± 0.00bc

0.00 ± 0.00c

0.00 ± 0.00b

1.52 ± 0.35bc

0.40 ± 0.14bc

0.00 ± 0.00b

1.36 ± 0.33b

1.52 ± 0.21b

10.21 ± 3.90b

Significance
-------------------------------------------------------P-value---------------------------------------------------Mulch species
<0.01
<0.01
<0.01
(S)
Mulch
0.01
<0.01
0.03
management (M)
SxM
<0.01
<0.01
0.01
a
Different letters within column indicate significant differences within main effects according to LSMeans (α=0.05).
b
These represent average mulch heights within the sampling quadrats.
c

These represent total mulch biomass within the sampling quadrats.

71

Table 3.2. Mean ± SEM mulch biomass at an early, intermediate, and late date during the growing season.
2 a,c

Main effects
Mulch species
Bare soil
Hairy vetch
Winter rye
Mulch management
No crimping
Crimped

Significance
Mulch species (S)
Mulch management (M)
SxM

Early season

Mulch biomass (g/m )
Intermediate
Season

0.00 ± 0.00a
23.91 ± 4.19b
47.53 ± 9.17c

0.00 ± 0.00a
1.72 ± 0.44b
36.16 ± 6.41c

0.00 ± 0.00a
2.32 ± 1.07b
41.35 ± 8.80c

24.84 ± 7.46
22.79 ± 7.46

10.69 ± 4.35
14.56 ± 6.95

15.00 ± 7.89
14.11 ± 6.04

Late season

----------------------------------------------------P-value-----------------------------------------<0.01
<0.01
<0.01
0.62
0.47
0.61
0.94
0.68
0.36

a
Different letters within column indicate significant differences within main effects according to LSMeans (α=0.05).
b
These represent average mulch heights within the sampling quadrats.
c

These represent total mulch biomass within the sampling quadrats.

72

Table 3.3. Mean ± SEM weed height, weed abundance, and weed species richness across mulch species and
management in 2012.
a
Weed abundance
Weed species richness
Weed height (cm)
2 a
2 a
(per m )
(per m )
Main effects
Mulch species
Bare soil
6.23 ± 1.04a
12.83 ± 2.39a
3.75 ± 0.31a
Hairy vetch
5.45 ± 0.92a
9.83 ± 1.50ab
3.38 ± 0.36ab
Winter rye
3.25 ± 0.67b
4.67 ± 0.83b
2.50 ± 0.32b
Mulch management
3.28 ± 0.29
No crimping
4.98 ± 0.77
8.14 ± 1.06
Crimped
4.98 ± 0.73
10.08 ± 1.80
3.14 ± 0.28
Significance
Mulch species (S)
Mulch management (M)
SxM

-------------------------------------P-value--------------------------------------< 0.01
< 0.01
0.02
0.99
0.29
0.67
0.53
0.26
0.39

a

Different letters within column indicate significant differences within main effects according to LSMeans
(α=0.05).

73

Table 3.4. Mean ± SEM non-marketable and marketable fresh weight of cabbage
heads at harvest.
a
a
Main effects
Marketable (kg)
Non-marketable (kg)
Mulch species
Bare soil
Hairy vetch
Winter rye
Mulch management
No crimping
Crimped

31.47 ± 5.46a
30.25 ± 4.71a
15.54 ± 3.21b

17.20 ± 2.85a
17.50 ± 1.99a
28.25 ± 1.59b

25.92 ± 4.42
25.58 ± 4.05

20.77 ± 2.23
21.20 ± 2.45

Significance
---------------------P-value-----------------------Mulch species (S)
0.03
<0.01
Mulch management (M)
0.94
0.93
SxM
0.12
0.44
a

Different letters within column indicate significant differences within main
effects according to LSMeans (α=0.05).

74

Table 3.5. Herbivore abundance in experimental plots across fixed effect: mulch management (no crimping or
crimped).
b
Source of variation
Insect abundance (Mean ± SEM)
No crimping plots

Crimping plots

Mulch management
F
P
df
a

Major Herbivores
c,e,h,i
Imported cabbageworm
Cabbage looper
Cabbage aphid

c,e,g,h

c,e,f,h

d,f

Crucifer flea beetles
Striped flea beetles
Onion thrips

d

d,f

d,e

16.64 ± 0.91

1,19

1.27

0.27

1.03 ± 0.11

1.11 ± 0.11

1,20

0.02

0.89

1.58 ± 0.14

1.73 ± 0.17

1,22

0.17

0.68

37.53 ± 2.07

Diamondback moth

16.93 ± 0.96

40.47 ± 2.79

1,22

0.55

0.47

18.75 ± 1.89

28.94 ± 3.93

1,21

0.82

0.38

14.71 ± 0.93a

12.97 ± 0.93b

1,22

6.89

0.02

1,19

0.04

0.84

130.81 ± 8.02

131.14 ± 7.60

a
Denominator degrees of freedom were calculated with the Kenward-Roger method (SAS 9.2, 2002-2008).
b
Different letters indicate significant differences within row according to LSMeans (α=0.05).
c
Abundance determined from visual abundance per plot.
d
Abundance determined from sticky trap abundance per plot.
e
Abundance determined using model with weed height as a covariate
f
Abundance determined using model with weed diversity as a covariate
g
Abundance determined using model with marketable cabbage yield as a covariate.
h
Abundance determined using model with total predator abundance as a covariate.
I

Abundance determined using model with total parasitoid abundance as a covariate.

75

Table 3.6. Natural enemy abundance in experimental plots across fixed effect: mulch
management (no crimping or crimped).
Insect abundance
Source of
b
variation
(Mean ± SEM)
No crimping
Crimping
Mulch
management
Major Natural Enemies
Df a
F
P
c,g
1.21 ± 0.18
1.06 ± 0.16
1,20 0.75 0.40
Spined soldier bug
d, f
2.58 ± 0.20
2.66 ± 0.26
1,20 0.12 0.73
Pink lady beetle
d
0.09 ± 0.02a
0.25 ± 0.04b
1,21 6.55 0.02
14 spotted lady beetle
d
0.12 ± 0.02
0.12 ± 0.02
1,15 0.00 1.00
Convergent lady beetle
d,g
4.25 ± 0.34
4.10 ± 0.36
1,25 0.68 0.42
Minute pirate bug
d
5.06 ± 0.32
5.16 ± 0.36
1,18 0.08 0.78
Predatory thrips
e,f,l
1.81 ± 0.27
1.54 ± 0.23
1,23 1.76 0.20
B. quadrimaculatum
e,g,l
0.58 ± 0.11
0.43 ± 0.09
1,21 0.15 0.71
Poecilus sp.
e,f,l
0.42 ± 0.11
0.21 ± 0.05
1,20 0.68 0.42
Poecilus chalcites
e,g,h
0.15 ± 0.05
0.17 ± 0.04
1,17 0.91 0.35
Cyclotrachelus sodalis
e,l
0.33 ± 0.09
0.32 ± 0.07
1,15 0.35 0.56
Scarites quadriceps
e,h,l
0.19 ± 0.06
0.05 ± 0.01
1,15 1.47 0.24
A. sanctaecrucis
l
2.65 ± 0.40
3.11 ± 0.47
1,19 1.51 0.23
Harpalus pennsylvanicus
e
0.10 ± 0.03
0.11 ± 0.03
1,17 0.00 0.97
Harpalus erraticus
e,l
0.10 ± 0.02
0.11 ± 0.03
1,16 0.01 0.93
Amara sp.
e,g
0.12 ± 0.03
0.12 ± 0.04
1,18 0.17 0.69
Pterostichus permundus
e,l
0.05 ± 0.02
0.24 ± 0.08
1,18 1.05 0.32
Cicindela sp.
e,l
0.26 ± 0.05
0.29 ± 0.06
1,19 0.02 0.88
Staphylinidae
e,l
0.05 ± 0.02
0.07 ± 0.02
1,20 1.42 0.25
Centipede
e,g,l
2.14 ± 0.21
2.25 ± 0.24
1,24 1.14 0.30
Lycosidae
e
0.23 ± 0.06
0.18 ± 0.04
1,18 0.02 0.90
Linyphidae
e,f,l
3.30 ± 0.57
3.67 ± 0.66
1,23 1.06 0.31
Phalangidae
d,f,j
6.30 ± 0.93
7.10 ± 0.49
1,19 0.02 0.88
Cotesia rubecula
d
0.63 ± 0.06
0.82 ± 0.14
1,20 0.04 0.85
Diadegma insulare
d,k
46.44 ± 1.92
49.1 ± 3.13
1,24 0.16 0.69
Chalcidoidea
a

Denominator degrees of freedom were calculated with the Kenward-Roger method
(SAS 9.2, 2002-2008).
b
Different letters indicate significant differences within row according to LSMeans
(α=0.05).
c
Abundance determined from visual abundance per plot.
d
Abundance determined from sticky trap abundance per plot.
e
Abundance determined from pitfall trap abundance per plot.
f
Abundance determined using model with weed height as a covariate.

76

Table 3.6 (cont’d)
g
Abundance determined using model with weed diversity as covariate.
h
Abundance determined using model with marketable cabbage yield as a covariate.
i
Abundance determined using model with total herbivore abundance as a covariate.
j

Abundance determined using model with imported cabbageworm abundance as a
covariate.
k
Abundance determined using model with total predator abundance as a covariate.

77

Table 3.7. Herbivore abundance in experimental plots across fixed effect: mulch species (bare soil, hairy vetch, rye).
b
Source of variation
Insect abundance (Mean ± SEM)
Bare soil

Diamondback moth
Cabbage looper
Cabbage aphid

c,e,g,h

c,e,f,h

d,f

Crucifer flea beetles
Striped flea beetles
Onion thrips

d

d,f

d,e

Rye

15.81 ± 0.37a

15.74 ± 1.15a

18.81 ± 1.20b

2,22

5.72

0.01

1.03 ± 0.12

1.24 ± 0.16

0.94 ± 0.10

2,20

1.01

0.38

1.66 ± 0.18

1.52 ± 0.17

1.78 ± 0.22

2,24

2.19

0.13

35.14 ± 2.87

Major Herbivores
c,e,h,i
Imported cabbageworm

Hairy vetch

Mulch species
a
F
P
df

41.38 ± 3.21

40.61 ± 2.96

2,63

1.51

0.22

31.53 ± 5.13

29.41 ± 3.63

10.05 ± 0.95

2,22

0.10

0.91

14.50 ± 1.27

14.04 ± 1.15

12.93 ± 1.00

2,25

1.47

0.25

124.71 ± 9.33

144.41 ± 10.84

123.61 ± 8.21

2,19

3.58

0.05

a
Denominator degrees of freedom were calculated with the Kenward-Roger method (SAS 9.2, 2002-2008).
b
Different letters indicate significant differences within row according to LSMeans (α=0.05).
c
Abundance determined from visual abundance per plot.
d
Abundance determined from sticky trap abundance per plot.
e
Abundance determined using model with weed height as a covariate
f
Abundance determined using model with weed diversity as a covariate
g
Abundance determined using model with marketable cabbage yield as a covariate.
h
Abundance determined using model with total predator abundance as a covariate.
i

Abundance determined using model with total parasitoid abundance as a covariate.

78

Table 3.8. Natural enemy abundance in experimental plots across fixed effect: mulch species (bare soil, hairy
vetch, rye).
b
Source of variation
Insect abundance (Mean ± SEM)
Bare soil

Hairy vetch

Rye

1.19 ± 0.20

1.14 ± 0.20

1.08 ± 0.21

2,82

2.27

0.11

2.61 ± 0.25a

2.99 ± 0.37a

2.25 ± 0.18b

2,27

4.59

0.02

d,f,g

0.21 ± 0.04a

0.23 ± 0.04a

0.07 ± 0.02b

2,20

3.06

0.07

d

0.13 ± 0.03

0.11 ± 0.03

0.11 ± 0.03

2,15

0.01

0.99

4.70 ± 0.45

4.87 ± 0.50

2.89 ± 0.28

2,26

2.51

0.10

4.51 ± 0.40

4.62 ± 0.33

6.25 ± 0.50

2,18

3.50

0.05

1.46 ± 0.29

1.89 ± 0.33

1.65 ± 0.30

2,90

1.74

0.18

0.95 ± 0.19a

0.44 ± 0.09b

0.13 ± 0.04c

2,23

14.45 <0.01

0.20 ± 0.07a

0.25 ± 0.08a

0.51 ± 0.14b

2,22

1.94

0.06

0.09 ± 0.04

0.18 ± 0.06

0.22 ± 0.06

2,17

1.00

0.39

0.39 ± 0.11

0.25 ± 0.07

0.35 ± 0.11

2,15

0.25

0.78

0.16 ± 0.08

0.11 ± 0.04

0.09 ± 0.04

2,15

0.14

0.87

1.13 ± 0.20a

2.07 ± 0.32b

5.51 ± 0.81c

2,22

23.91 <0.01

0.04 ± 0.02

0.09 ± 0.04

0.16 ± 0.04

2,17

2.15

0.15

0.07 ± 0.02

0.09 ± 0.04

0.15 ± 0.04

2,16

2.62

0.10

0.05 ± 0.03a

0.02 ± 0.03a

0.29 ± 0.06b

2,18

17.85 <0.01

e,l

0.09 ± 0.04

0.16 ± 0.08

0.18 ± 0.10

2,18

0.01

0.99

e,l

0.18 ± 0.06

0.40 ± 0.08

0.24 ± 0.06

2,19

2.93

0.08

Major Natural Enemies
Spined soldier bug

c,g

d, f

Pink lady beetle

14 spotted lady beetle

Convergent lady beetle
Minute pirate bug
Predatory thrips

d,g

d

B. quadrimaculatum
Poecilus sp.

e,f,l

e,f,l

Poecilus chalcites

e,f,l

Cyclotrachelus sodalis
Scarites quadriceps
A. sanctaecrucis

e,g,h

e,l

e,h,l

Harpalus pennsylvanicus
Harpalus erraticus
Amara sp.

l

e

e,l

Pterostichus permundus
Cicindela sp.

Staphylinidae

e,g

79

Mulch species
a
F
P
df

Table 3.8 (cont’d)
b

Insect abundance (Mean ± SEM)
Bare soil
Hairy vetch
Rye

Source of variation

e,l

0.11 ± 0.04

0.02 ± 0.01

0.05 ± 0.03

2,20

1.47

0.25

e,g,l

2.14 ± 0.29

2.47 ± 0.25

1.96 ± 0.27

2,25

2.07

0.15

e

0.07 ± 0.02a

0.18 ± 0.05a

0.36 ± 0.08b

2,18

10.71 <0.01

3.50 ± 0.71

3.60 ± 0.76

3.35 ± 0.79

2,23

1.83

0.18

4.89 ± 0.88a

6.54 ± 1.08a

8.78 ± 1.82b

2,15

4.62

0.03

0.59 ± 0.06

0.84 ± 0.16

0.76 ± 0.16

2,21

0.20

0.82

51.60 ± 4.53

45.44 ± 2.03

46.11 ± 2.28

2,24

0.88

0.42

Major Natural Enemies
Centipede
Lycosidae

Linyphidae

Phalangidae

e,f,l

Cotesia rubecula

d,f,j

Diadegma insulare
Chalcidoidea

d,k

d

Mulch species
a
F
P
df

a
Denominator degrees of freedom were calculated with the Kenward-Roger method (SAS 9.2, 2002-2008).
b
Different letters indicate significant differences within row according to LSMeans (α=0.05).
c
Abundance determined from visual abundance per plot.
d
Abundance determined from sticky trap abundance per plot.
e
Abundance determined from pitfall trap abundance per plot.
f
Abundance determined using model with weed height as a covariate.
g
Abundance determined using model with weed diversity as covariate.
h
Abundance determined using model with marketable cabbage yield as a covariate.
i
Abundance determined using model with total herbivore abundance as a covariate.
j
Abundance determined using model with imported cabbageworm abundance as a covariate.
k

Abundance determined using model with total predator abundance as a covariate.

80

Total predators per cabbage plot

40

Bare soil, no crimping
Vetch, no crimping
Rye, no crimping
Bare soil, crimping
Vetch, crimping
Rye, crimping

35
30
25

∆

*

20

∆

*
*

15

∆

10
5
0
6/18

6/25

7/2

7/9

7/16

7/23

7/30

8/6

8/13

8/20

8/27

Figure 3.1. Total predator abundance on yellow sticky traps throughout the cabbage growing season in experimental plots with
different mulch species (bare soil, rye, vetch) and mulch management treatments (no crimping, crimping). Asterisks and delta symbols
above lines indicate significant differences across mulch species and management, respectively according to LSMeans (α= 0.5).

81

Total ground dwelling
arthropods per cabbage plot

60
Bare soil, no crimping
Vetch, no crimping
Rye, no crimping
Bare soil, crimping
Vetch, crimping
Rye, crimping

50
40
30
20

∆*
10
0
6/13

6/20

6/27

7/4

7/11

7/18

7/25

8/1

8/8

8/15

8/22

8/29

Figure 3.2. Total ground dwelling predators throughout the cabbage growing season in experimental plots with different mulch
species (bare soil, rye, vetch) and mulch management treatments (no crimping, crimping). Asterisks and delta symbols above lines
indicate significant differences across mulch species and management, respectively according to LSMeans (α= 0.5).

82

Total parasitoids per cabbage plot

250
Bare soil, no crimping
Vetch, no crimping
Rye, no crimping

200

Bare soil, crimping
Vetch, crimping
Rye, crimping

150
100

*
*

50
0
6/18

6/25

7/2

7/9

7/16

7/23

7/30

8/6

8/13

8/20

8/27

Figure 3.3. Total parasitoid abundance on yellow sticky traps throughout the cabbage growing season in experimental plots with
different mulch species (bare soil, rye, vetch) and mulch management treatments (no crimping, crimping). Asterisks above lines
indicate significant differences across mulch species according to LSMeans (α = 0.5).

83

% Larval mortality

80
70
60
50
40
30
20
10
0

A

Early season
Intermediate season

A

A

A

A

A

A

A

A

Late season

B

B

No see um

B

Mosquito

Sham

No net

Figure 3.4. Percent larval mortality (Mean ± SEM) of the imported cabbageworm across natural enemy exclusion cages. Letters above
bars indicate significant differences according to LSMeans (α = 0.5).

84

% Larvae parasitized

35

A

30

Early season

25

Intermediate season

20

A
A

Late season

15

A

10
5

A

A

A

B

B

AB

AB

B

0
No see um

Mosquito

Sham

No net

Figure 3.5. Percent of imported cabbageworm larvae parasitized (Mean ± SEM) across natural enemy exclusion cages. Letters above
bars indicate significant differences according to LSMeans (α = 0.5).

85

Discussion
Mulch, weed, and cabbage yield measurements. Mulch species and management or rowmiddles created habitats within the cabbage field with distinct structural complexity and
vegetational complexity (e.g. weed community). For example, mulch management changed the
height of the mulch in the row middles, with crimping reducing mulch height during the early
and intermediate portion of the growing season (Table 3.1). On the other hand, different mulch
species resulted in different amounts of biomass present in the row middle with rye plots
containing the greatest amount of structure. Hairy vetch mulch decayed quicker than rye mulch,
leading to less structure in the row middles. These plots resembled bare soil plots late in the
growing season. Additionally, rye mulch successfully suppressed weed height, abundance, and
richness compared to bare soil plots (Table 3.3). This supports past observations in vegetable
crops by Liebman and Dyck (1993) and Teasdale (1998).
The presence of rye mulch reduced marketable cabbage yield and led to greater nonmarketable yield (Table 3.4). Negative effects of rye on cabbage yield have been observed in the
past and may be the result of rye’s allelopathic effects, light competition, or tie-up of nitrogen
from slow decaying rye (Schonbeck et al. 1993). It is suggested that killing the rye cover crop
earlier in the season or reducing the density of the cover crop may reduce these effects and
maintain cabbage yields comparable to non-mulched plots (Mochizuki et al. 2008). It should also
be noted that no nitrogen sidedress application was made, so nitrogen was likely limiting,
especially in rye treatments.

Arthropod communities. The response of mobile predators sampled on sticky traps was
inconsistent during the growing season. Total predator abundance responded negatively to

86

mulching during the early portion of the cabbage growing season and total predator abundance
was greater in mulched plots later in the season (Figure 3.1). Mulch management also appeared
to have a mixed effect on total predator abundance. When considering abundance of ground
dwelling predators, ground cover had little effect on total predator abundance (Figure 3.2).
However, on the last sampling date, ground dwelling predators appeared to prefer rye plots
without crimping.
Positive responses of predators to mulching have been observed in potatoes, peppers, and
collards (Johnson et al. 2004, Pullaro et al. 2006). In these cases, mulching may create habitats
that provide additional food sources (e.g. detritivore prey) and protection (Langellotto and Denno
2004). In addition to this study, negative responses of predators to mulching have also been
observed in crops such as potatoes and turnips (Wiedenmann et al. 2004, Manrique et al. 2010).
Complex habitats may serve to interfere with visual and chemical cues (e.g. Finch and Collier
2000, Randkofer et al. 2010).
During the early cabbage growing season, parasitoid abundance was positively impacted
by the presence of rye mulch (Figure 3.3). However, parasitoid populations and parasitism rates
of the imported cabbageworm (Figure 3.5) are minimal during the early season and therefore
may have a limited impact on pest suppression.

Arthropod abundance. Mulch management and species had an impact on two pests: the striped
flea beetle and the imported cabbageworm (Table 3.5). In the past, vegetational complexity
provided by Brassica weed species in collard plots had an impact on flea beetle populations
(Schellhorn and Sork 1997), with more flea beetles present in plots with weed polycultures than
bare soil. This study found structural complexity to influence flea beetle populations. In

87

particular, striped flea beetles were slightly more abundant in standing rye mulched cabbage
plots compared to crimped plots. It appears that the height of the mulch did not interfere with
pest movement or host finding into plots with standing rye, although the height of surrounding
companion plants has demonstrated an interference effect in the past with another Brassica pest,
the cabbage root fly (Finch et al. 2003).
The imported cabbageworm responded to mulch species, with greater abundance in plots
with rye mulch compared to bare soil or vetch plots. This is contrary to past observations of the
imported cabbageworm, in which populations were lower in cabbage with rye mulch than non
mulched plots (Bottenberg et al. 1997). Banks (1998) has attributed any difference in imported
cabbageworm abundance across habitats of varying complexity with differences in predation. He
suspected that the imported cabbageworm was unresponsive to within field complexity due to
the high mobility, flight patterns, and widespread oviposition patterns of the imported
cabbageworm. Instead, the greater structure in the row middle may provide shelter and greater
protection from predators.
Mulch management and species impacted the abundance of several natural enemies
(Table 3.6). The presence of rye mulch was associated with lower populations of the pink lady
beetle and Poecilus sp. ground beetle. For the pink lady beetle, these results are contrary to the
findings of Szendrei and Weber (2009), who observed greater abundance of the pink lady beetle
in mulched potatoes compared to control potato plots. Reduced predator populations suggest a
reduced encounter rate and hunting efficiency of predators in complex agricultural environments
(Legrand and Barbosa 2003).
This study found a positive impact of rye mulch on the abundance of the generalist
predators: Harpalus pennsylvanicus, Pterostichus permundus, and Linyphiidae spiders (Table

88

3.8). Contrary to these results, Wiedenmann et al. (2004) found no impact of mulching on the
abundance of ground beetle species when comparing mulched and non-mulched garden plots.
Additionally, a review by Sunderland and Samu (2000) reported that in previous studies,
mulching had no effect on Linyphiid spider abundance compared to other forms of habitat
management.
Several mechanisms may explain these positive responses of generalist predator
abundance to habitat complexity. Elton (1958) first proposed the enemies hypothesis, suggesting
complex habitats would enhance predation of pests in cabbage. Complex habitats may provide
protection, further resources, ideal microclimates, or alternative prey such as detritivores which
may support greater natural enemy populations (Root 1973, Landis et al. 2000, Sunderland and
Samu 2000). These complex habitats may also increase the availability of prey or hosts, or
enhance the efficiency of natural enemies. Complex habitats may reduce negative interactions
between predators, including intraguild predation such as cannibalism, allowing greater pest
suppression (Finke and Denno 2002, Finke and Denno 2006).
The specialist parasitoid Cotesia rubecula, was significantly more abundant in rye
mulched plots (Table 3.8). Vegetational complexity has influenced C. rubecula abundance in the
past with higher populations in broccoli with interplanted flowers (Zhao et al. 1992). The impact
of structural complexity on C. rubecula provided by rye and vetch mulch has not been reported
in current literature. Cotesia rubecula species relies on specific odor cues from the feeding
damage of its host, the imported cabbageworm. The parasitoid will preferentially land on and
search plants with imported cabbageworm feeding damage (Agelopoulos and Keller 1994b).
However, it is important to note that mulching did not interfere with these signals.

89

Significant covariates for herbivores. Weed height or diversity impacted the abundance
of every major herbivore. This supports results from 2010 and 2011, in which weed pressures
influenced herbivore abundance. I suspect that weeds provide shelter, alternative resources, and
attractive microclimates for these pest species. In general, total predator abundance also
impacted abundance of the imported cabbageworm, diamondback moth, and cabbage looper,
suggesting efficiency of predators and predation may be influencing herbivore populations in the
field.
Significant covariates for natural enemies. Weed height or diversity also impacted
natural enemy abundance for eight major natural enemies. The relationships between weeds and
arthropod abundance in Brassica crops has been demonstrated in the past, including the role of
weed height (Finch et al. 2003) and weed diversity (Andow 1988, Schellhorn and Sork 1997,
Broad et al. 2008). The weeds in the row-middle may be providing shelter, resources, alternative
prey. Alternatively, weeds may reduce time spent hunting on cabbage plants or influence hunting
efficiency leading to changes in natural enemy abundance. Marketable cabbage yield had a
limited impact on predators, with a single ground beetle responding negatively to marketable
cabbage yield. In this case, Cyclotracehlus sodalist may find hunting and movement easier on
smaller cabbage plants. Although the generalist predators in this study did not appear to be
influenced by prey abundance, host abundance influenced C. rubecula and Chalcidoidea
abundance. As in 2010 and 2011 (see Chapter 2), Cotesia rubecula and the hyperparasitoids
included in this study appear to be following their hosts and potentially the stronger host odor
signals in mulched plots. Total predator abundance significantly impacted multiple ground
dwelling arthropods. For the predators negatively impacted by predator abundance (e.g. Poecilus
sp.), intraguild predation may be impacting species abundance.

90

Natural enemy exclusion cages. Mortality and parasitism rates in the cabbage field were not
impacted by mulch management or species. Similar results have been observed in potatoes, in
which mulching had no impact on predation rates (Szendrei et al. 2010). Although natural enemy
abundance was greater in rye mulch plots, the presence of cover crop mulch may interfere with
the mobility of predators searching for prey, reducing the efficiency of predators (Legrand and
Barbosa 2003, Tylianakis and Romo 2010). Complex habitats may interfere with visual (Ferran
and Dixon 1993) or chemical (Randlkofer et al. 2010) searching cues used by natural enemies,
leading to lower capture and consumption rates. Intraguild predation, including cannibalism, may
also be greater in complex habitats with higher predator populations (Tylianakis and Romo
2010).
The results from the natural enemy exclusion cages highlight the importance of small
predators in cabbage pest management. At each sampling date, the guild of small predators was
responsible for the majority of predation within the cages. In other studies, larger natural enemies
(e.g. lady beetle adults, large ground beetles, etc.) are often surveyed and examined for their pest
suppression in cabbage (e.g. Grevstad and Klepetka 1992, Armstrong and McKinlay 1997),
while the guild of small natural enemies is overlooked. Natural enemies such as syrphid fly
larvae, lacewing larvae, coccinellid larvae, minute pirate bugs, and Linyphiid spiders which were
observed feeding on prey inside the exclusion cages, have provided biological control in past
studies (Eigenbrode et al. 1996, Schmaedick and Shelton 2000, Reddy et al. 2004). Many of
these natural enemies have demonstrated an ability to readily consume cabbage pests and search
waxy leaf surfaces that larger predators such as coccinellid adults, Lycosids, and Phalangiids
cannot access (Eigenbrode et al. 1996, Schmaedick and Shelton 2000, Reddy et al. 2004). Future

91

research in conservation biological control with cabbage should examine practices that enhance
this guild of small predators.

Summary. Mulch management and species had no impact on most pest species. The presence
of mulch had mixed effects on natural enemies, although several ground beetles and Linyphiid
spiders were positively impacted by mulching. The guild of small predators including these
Linyphiid spiders, appear to be responsible for most predation in the field.
In addition to the direct effects of the mulching (Figure 1.3), possible indirect effects of
mulching (Figure 1.4) were considered through the examination of covariates such as marketable
cabbage yield, herbivore abundance, and natural enemy abundance. Although the exact
mechanisms for the trends observed in arthropod abundance are unknown, it appears that both
direct and indirect factors are influencing arthropod populations, particularly weed height and
diversity.
It is possible that habitat structure at a larger spatial scale would have had a larger impact
on insect abundance and biological control. At larger spatial scales, trap cropping, floral borders,
and diversity at the landscape level have been shown to impact natural enemy abundance (e.g.
Mitchell et al. 1997, Mitchell et al. 2000, Hooks and Johnson 2003, Lee and Heimpel 2005,
Jonsson et al. 2012). On a smaller scale, changes in leaf surface composition (Eigenbrode et al.
1996) and plant architecture (Gingras et al. 2003) can also modify natural enemy and pest
abundance, although this was not examined in my study.
Although a diverse community of pests attack cabbage, it is necessary to examine the
impact of mulching on predation and parasitism of other pests, particularly other lepidopteran
pests causing significant damage in Michigan cabbage. Future studies should focus on mulching

92

practices that maintain cabbage yields while enhancing predation and parasitism of several
cabbage pests. The mechanisms behind the trends observed in this study are still unclear. In
particular, future work should determine whether visual or odor cues are involved in the habitat
preferences of those natural enemies that were positively influenced by the presence of mulches.

93

Ch. 4
The role of olfactory cues from cover crop mulch on
two generalist and one specialist natural enemy in cabbage

Introduction. The imported cabbageworm Pieris rapae L. (Lepidoptera: Pieridae) is considered
one of the greatest foliar cabbage pest in the United States (Schmaedick and Shelton 1999,
Wold-Burkness et al. 2005). In Michigan, this pest is often one of the first butterflies to appear in
the spring and last butterflies observed in the fall. The presence of this pest throughout the
cabbage growing season, its high fecundity, and its high mobility (e.g. Richards 1940), have led
the imported cabbageworm to become a serious threat to both early and late season cabbage
production in Michigan.
Natural enemies can provide significant control of the imported cabbageworm in cabbage
fields. Schmaedick and Shelton (1999) observed up to 80% egg and larval mortality due to
natural enemies. Lady beetles (Coleoptera: Coccinellidae), predatory stink bugs (Hemiptera:
Pentatomidae), spiders (Araneae), and ground beetles (Coleoptera: Carabidae) are among the
generalist predators feeding on the imported cabbageworm (Schellhorn and Sork 1997,
Schmaedick and Shelton 2000, Suenaga and Hamamura 2001, Moreno et al. 2010). Specialist
natural enemies such as the parasitoid Cotesia rubecula Marshall (Hymenoptera: Braconidae),
provide additional pest suppression. In Michigan, parasitism rates of the imported cabbageworm
by C. rubecula have recently been observed up to 71% (Herlihy et al. 2012).
Through the manipulation of the habitat within a cabbage field, these natural enemies
may be conserved, enhancing the pest management services they provide. Past studies have
observed habitats with greater structural complexity to support greater natural enemy populations.

94

Although enhancement of natural enemy populations has been observed through habitat
management in a variety of crops (Landis et al. 2000), it is unclear how different forms of
structural complexity due to mulching may influence the natural enemy community in cabbage.
Xu et al. (2011) observed greater natural enemy populations in the presence of mulch. However,
the response and strength of arthropod responses may vary with different types of mulching
practices.
For arthropods that respond positively to mulching, structurally complex habitats may
provide alternative prey, enhance efficiency of predators, and provide protection from intraguild
predators (Langellotto and Denno 2004). It is also suspected that structurally complex habitats
influence the visual and chemical cues that natural enemies rely on when searching for prey and
hosts in cabbage (e.g. Finch and Collier 2000, Randkofer et al. 2010). Chemical cues appear to
play a particularly important role in attraction of natural enemies to the habitat and ability of the
natural enemy to locate herbivores within a cabbage field.
Feeding damage and frass from Pieris rapae larvae release a blend of volatiles into the
surrounding environment (Agelopoulos and Keller 1994c, Mumm et al. 2008). These “resourceindicating odors” indicate the presence of P. rapae and have influenced the behavior of natural
enemies in the past (Schroeder and Hilker 2008). Generalist predators, such as the spined soldier
bug, Podisus maculiventris (Hemiptera: Pentatomidae), have demonstrated less discrimination of
odor cues than specialist natural enemies. The responsiveness of generalist predators to damaged
cabbage may be limited by cabbage cultivar and environmental conditions (Vuorinen et al. 2004).
Generalists may also respond to a variety of resource-indicating odors from multiple herbivores
or they may not rely primarily on olfactory cues to find herbivores (Rossbach et al. 2005).

95

On the other hand, the specialist Cotesia rubecula has demonstrated consistent attraction
to the odors from cabbage damaged by caterpillars, including the diamondback moth and the
imported cabbageworm (van Poecke et al. 2003). Cotesia rubecula is also highly attracted to the
frass of imported cabbageworm larvae (Agelopoulos and Keller 1994a). The presence of these
volatiles resulting from larval feeding lead C. rubecula to land on infested cabbage plants and
increases the parasitoid’s time spent searching for its host (Agelopoulos and Keller 1994b). This
suggests C. rubecula relies on relatively specific odor cues from lepidopteran damage compared
to generalist predators in cabbage.
Complex habitats create a greater diversity of background odors in the environment. This
complex blend of odors accompanies the more specific cues from the herbivores that indicate
their presence. Schroeder and Hilker (2008) have found background odors to be either: 1)
irrelevant to the natural enemy’s ability to find herbivores 2) interfere with the natural enemy’s
ability to find herbivores due to masking of the resource-indicating odor or 3) enhance a natural
enemy’s ability to find herbivores. It is unknown how background odors from decaying cover
crop mulches influence natural enemies. However, it is possible that discrimination of odors
from different mulching treatments may also vary between generalist and specialist natural
enemies.
In this study, I examined the effects of mulch species and management on arthropod
behavior. I observed the role of odor cues from mulch on habitat choices of two generalist
predators (spined soldier bug and convergent lady beetle) and one specialist parasitoid (Cotesia
rubecula).

96

Materials and Methods
Y-tube olfactometer
Insects.Imported cabbageworm larvae were reared at Michigan State University in a
greenhouse colony. Imported cabbageworm originated from the MSU Horticulture Farm (East
Lansing, MI) and maintained for c.a. 10 generations. Insects were kept in 47.5 x 47.5 x 93 cm
mesh bug dorm cages (Megaview Science Co., Ltd., Taichung, Taiwan) with a 16:8h L:D cycle
at 23-28°C. Larvae were reared on Brussels sprouts (Brassica oleracea var. “Churchill”;
Johnny’s Selected Seeds, Albion, ME) ad libitum and adults were fed a 10% honey solution
upon emergence (Pure Clover Honey, Meijer Distribution Inc., Lansing, MI).
Third instar spined soldier bugs were provided by T.A. Coudron (Biological Control of
Insects Research Laboratory, USDA-Agricultural Research Service, Columbia MO). For rearing
conditions and diet of spined soldier bugs, see Coudron et al. (2012). These predators were used
for assays upon arrival.
Adult convergent lady beetles were purchased from Rincon-Vitova Insectaries (Ventura,
CA). Beetles were placed in petri dishes (150 x 15 mm; VWR, Radnor, PA) and fed a 10%
honey solution upon arrival. Convergent lady beetles were held at 10°C until needed for assays.
Convergent lady beetles were removed from the fridge 24 h prior to the start of an assay and kept
in the laboratory at 25°C. To eliminate any effects that the sex of the parasitoid may have on
odor preference, only female beetles were used in subsequent assays.
Cotesia rubecula were reared in a greenhouse colony at Michigan State University.
Cotesia rubecula were originally collected from the MSU Horticulture Farm in summer 2012
and maintained c.a. 6 generations. Parasitoids were reared on imported cabbageworm from the

97

greenhouse colony previously described. Adult wasps were fed a 10% honey solution ad libitum
upon emergence. Due to the unresponsiveness of naïve parasitoids, adult parasitoids were
exposed to first and second instar imported cabbageworm for 24 h prior to use in y-tube assays.
Only female wasps were used in assays.

Odor sources. Odor sources included clean air, mulch (vetch, rye, or both vetch and rye),
damaged cabbage, and damaged cabbage with different mulches (vetch, rye, or vetch and rye).
Among these odor sources, ten treatments were tested within the y-tube olfactometer (see Table
4.1 for list of treatments).
Each odor source consisted of an outer pot (14cm diameter x 12cm height) with a smaller
inner pot (8cm x 9cm) placed inside (Figure 4.1). The tops of the two pots were flush with one
another. Both pots were filled with plant medium (Sure-Mix Perlite, Michigan Grower Products,
Galesburg, MI). For the clean air odor source, both the inner and outer pot was filled with plant
medium. For odor sources with damaged cabbage, the inner pot contained three cabbage plants.
Cabbage seedlings (Brassica oleracea var. “Farao”; Johnny’s Selected Seeds, Albion, ME) were
grown for one month (16:8h L:D; 23-28°C) before transplanting three cabbage plants into a 8cm
x 9cm pot together. Cabbage plants were grown separately and maintained in separate pots from
mulch species to prevent any confounding effects or interactions between mulch and cabbage
plants.
For odor sources with mulch, winter rye (Secale cereale; Dirt Works, New Haven, VT),
hairy vetch (Vicia villosa; Johnny’s Selected Seeds, Albion, ME), or both species were planted in
the space between the inner and outer pots. Winter rye was planted at a rate of 0.25 ounce/pot.
Hairy vetch was planted at a rate of 0.13 ounce/pot. Mulch species were planted at these same

98

rates for odor sources containing both rye and vetch. After one month, RoundUp was applied to
all above ground foliage of cover crop plants and a week later, cover crop mulch was used in
bioassays.
Twenty four hours before an assay began, odor sources were placed in individual screen
cages (30 x 30 x 30 cm; Bioquip Products, Racho Dominguez, CA). At this time, all inner pots
were placed within outer pots. To create herbivore damage on cabbage, six imported
cabbageworm at the 5

th

instar stage were placed on the three cabbage plants in the inner pots and

allowed to feed for 24 h.
Table 4.1. List of treatments or choice tests performed for each natural enemy species
(spined soldier bug, convergent lady beetle, and Cotesia rubecula).
Option 1
Option 2
Air
Damaged cabbage
Air
Vetch mulch
Air
Rye mulch
Air
Vetch and rye mulch
Vetch mulch
Damaged cabbage
Rye mulch
Damaged cabbage
Vetch and rye mulch
Damaged cabbage
Damaged cabbage
Damaged cabbage and vetch mulch
Damaged cabbage
Damaged cabbage and rye mulch
Damaged cabbage
Damaged cabbage and vetch mulch and rye mulch

99

A

B

C

D

E

F

G

H

Figure 4.1. Odor sources for y-tube olfactometer. Top row (left to right): clean air (A) and
damaged cabbage (B). Second row (left to right): vetch (C), rye (D), and both rye and vetch
mulch (E). Bottom row (left to right): cabbage with vetch (F), cabbage with rye (G), and cabbage
with both vetch and rye mulch (H).

Y-tube assay. To test olfactory preferences of the spined soldier bug, convergent lady
beetle, and Cotesia rubecula, a y-tube olfactometer was used (three arms: two 6 cm arms and one
12 cm arm, 2cm diameter, ground glass joints; Michigan State University Glass Blowing Facility,
East Lansing, MI). Air filtered through activated charcoal was divided into two streams, each
travelling in Teflon tubing leading to two glass chambers (7 L each; Corning Incorporated,
100

Corning, NY) containing different odor sources and attached to the olfactometer (Analytical
Research Systems, Gainesville, FL). Air flow was adjusted for each natural enemy species. Air
flow was maintained at 1 L/min for both adult convergent lady beetles and Cotesia rubecula
adults. However, air flow was reduced for spined soldier bugs and maintained at 0.50 L/min
(Vuorinen et al. 2004).
Individual natural enemies were placed at the end of the Y-tube and observed for 15
minutes or until a choice was made. If an insect did not respond within 15 minutes, it was
considered nonresponsive. Individual natural enemies were tested once and odor sources were
replaced after six insects were assayed. The position of the odor source (left or right chamber)
was placed at random and changed between replicates to eliminate positional bias. Between runs,
glassware was rinsed with methanol and hexane before drying at 60°C. Each treatment was
replicated 28 to 48 times for each natural enemy species.
For each odor source, the number of insects making a choice was analyzed. A G-test with
William’s correction (Sokal and Rohlf 1995) was performed to test the null hypothesis that the
natural enemy had no preference between odor sources.

Results
Y-tube olfactometer. The majority of natural enemies responded in the y-tube, in particular
86% of spined soldier bugs, 95% of convergent lady beetles, and 90% of C. rubecula made a
choice (Figure 4.2, 4.3, 4.4). However, the three natural enemies tested were not attracted to rye
mulch, vetch mulch, or both hairy vetch and rye mulch compared to clean air (Gadj < 3.47, df =
1, P > 0.06).

101

When compared to clean air, the spined soldier bug (Gadj = 0.91, df = 1, P = 0.34; Figure
4.2) and the convergent lady beetle (Gadj = 0.58, df = 1, P = 0.45; Figure 4.3) did not prefer
damaged cabbage. Additionally, the generalist predators were not attracted to damaged cabbage
compared to rye mulch, vetch mulch, or both rye and vetch mulch (Gadj < 1.38, df = 1, P > 0.24).
When damaged cabbage was offered opposite damaged cabbage with different mulch species
(e.g. rye, vetch, or rye and vetch mulch), the generalist predators were not repelled or attracted to
the presence of mulch (Gadj < 1.47, df = 1, P > 0.22).
Contrary to the generalist predators, the specialist natural enemy C. rubecula was
attracted to damaged cabbage. Significantly more C. rubecula chose damaged cabbage over
clean air (Gadj = 12.31, df = 1, P < 0.01; Figure 4.4). Damaged cabbage was also more attractive
to C. rubecula than rye mulch (Gadj = 21.05, df = 1, P < 0.01), vetch mulch (Gadj = 13.77, df = 1,
P < 0.01), or both mulch species combined (Gadj = 5.18, df = 1, P = 0.03). When the parasitoid
was offered damaged cabbage opposite damaged cabbage with different mulch species, it was
not affected by damaged cabbage with vetch mulch (Gadj = 0.31, df = 1, P = 0.58) or combined
rye and vetch mulch (Gadj = 1.27, df = 1, P = 0.26). However, it was repelled by damaged
cabbage with rye mulch present (Gadj = 10.45, df = 1, P < 0.01).

102

No choice
3

Air

5

Air

7

Rye

Air

10

Air

Vetch
Rye + Vetch

2
4

Cabbage

3
4

Rye

Cabbage

4

Vetch

Cabbage

Rye + Vetch

Cabbage

Cabbage + Rye

Cabbage

7
30

Cabbage

Cabbage + Vetch

Cabbage
20

Cabbage + Rye + Vetch
10

0

Figure 4.2. Response of spined soldier bug nymphs in Y-tube olfactometer choice test.

103

10

20

30

No choice
1

Air
Air

1

Air

3
1
0

Cabbage
Rye
Vetch
Rye + Vetch

Air
Cabbage

2

Cabbage

1

Rye

Cabbage
Cabbage

3
3

Rye + Vetch
Cabbage + Rye

Cabbage

3
30

Vetch

Cabbage + Vetch
Cabbage + Rye + Vetch

Cabbage
20

10

0

Figure 4.3. Response of convergent lady beetles in Y-tube olfactometer choice test.

104

10

20

30

No choice
4

Air

5

Cabbage *

Air

1

Rye
Vetch

Air

3

Air

1

Rye + Vetch

* Cabbage

4

Rye
Vetch

* Cabbage
* Cabbage

0
6

Rye + Vetch
Cabbage + Rye

* Cabbage

2

Cabbage

2

Cabbage + Vetch

Cabbage
40

30

20

Cabbage + Rye + Vetch
10

0

10

20

30

Figure 4.4. Response of Cotesia rubecula parasitoids in Y-tube olfactometer choice test. Asterisks indicate significant difference
according to G-test with William’s correction (α = 0.05).

105

Conclusions
Y-tube olfactometer. The choice tests indicated that generalist and specialist natural enemies
discriminate odor cues differently when choosing a habitat and searching for an herbivore.
Differential responses of generalist and specialist herbivores and natural enemies have been
observed in the past and are due to variability of glucosinolate concentrations in Brassica
cultivars (Giamoustaris and Mithen 1995, Li et al. 2000). Convergent lady beetles and spined
soldier bugs were not attracted to damaged cabbage. Although the spined soldier bug and
convergent lady beetle have preferred herbivore damaged plants in published studies (Hamilton
et al. 1999, Sant'Ana et al. 1999, Acar et al. 2001), this may be due to the inability of these
species to discriminate resource-indicating odors with this cabbage cultivar (Vuorinen et al.
2004) or their need for a boquet of resource-indicating odors from multiple herbivores (Rossbach
et al. 2005). These generalist predators may also rely on visual cues rather than odors to detect
appropriate habitat and landing sites (Finch and Collier 2000).
On the other hand, C. rubecula was highly attracted to resource-indicating odors from
damaged cabbage and appears to rely heavily on odor cues for host searching and location. The
ability of the imported cabbageworm feeding damage and frass to release attractive volatiles has
been well documented for natural enemies such as C. rubecula (e.g. Agelopoulos and Keller
1994a, c, van Poecke et al. 2003, Mumm et al. 2008).
The background odors of rye and vetch mulch were neither attractive nor repellent to the
generalist natural enemy species tested here; therefore mulch odors appear to be irrelevant
background odors in this study. The natural enemies tested may lack receptors to recognize the
volatiles from mulch or may become adapted to the mulch odor (Schroeder and Hilker 2008).
The lack of mulch odor preference may help to explain the similarity of convergent lady beetle

106

and spined soldier bug abundance across mulching treatments (see Chapter 3, Table 3.4, 3.6).
For the specialist, C. rubecula, mulch volatiles do not appear to interfere with the parasitoids
preference for damaged cabbage, with the exception of rye mulch. Rye mulch can be considered
a masking background odor (Schroeder and Hilker 2008). Rye did not elicit a repellent response
from C. rubecula when presented alone, however when rye is present with the host plant, it
reduced the parasitoids preference for damaged cabbage. Since C. rubecula abundance was
greater in cabbage field plots with rye mulch (see Chapter 3, Table 3.6), the masking effects of
the rye mulch on parasitoid behavior appear to be less important than other factors influencing
parasitoid behavior (e.g. intraguild predation, microclimates).

Summary. Mulching does not appear to attract natural enemies and mulching odors are not
repellent to generalist predators. However, odors from rye mulch may have masking effects for
the specialist parasitoid, Cotesia rubecula. Although rye mulch may mask cabbage damage odors,
this parasitoid was more abundant in experimental plots with rye mulch (see Chapter 3). Future
studies should examine the mechanisms behind the greater presence of C. rubecula in mulched
plots.

107

CHAPTER 5.
Conclusions and Future Research
Michigan cabbage production is threatened by a diverse community of insect pests.
Habitat management practices that manipulate habitat complexity can conserve natural enemies
and enhance biological control, providing improved management of this diverse pest community
in cabbage. Enhancing natural enemy populations by habitat management within cabbage fields
may reduce grower’s reliance on chemical controls, saving them time and money. These
practices also lead to greater sustainability and fewer negative impacts on the surrounding
environment. This thesis examined the relationship between cabbage arthropods and alternative
strategies that manipulate vegetational (living) and structural (non-living) complexity between
cabbage rows, in the hopes of providing additional management strategies for Michigan cabbage
growers.
In chapter two, I examined habitat management strategies such as reduced weed
management intensity and later oat cover crop kill date in strip-tilled cabbage. Greater
vegetational complexity from reduced weed management intensity led to reduced abundance of
several key cabbage pests. In general, greater vegetational and structural complexity between
cabbage rows enhanced populations of generalist natural enemies, including several generalist
predators with active hunting strategies. These results suggest reduced weed management and
mulch from late killed oats may enhance pest suppression through conservation of natural
enemies and could be incorporated into integrated pest management plans. However, before
these alternative habitat management strategies can be implemented, further work is needed:
ultimately, cabbage yields must be comparable to conventional practices. Future research should

108

fine-tune weed management and mulching strategies that maintain cabbage yields and do not
enhance hyperparasitism.
In chapter three, I examined different mulch species and mulch management. The goal of
this chapter was to determine the value of different mulching practices, including the use of
different mulch species (e.g. none, rye, vetch) and management (e.g. crimping or no crimping),
by examining the response of arthropods to these management methods. In general, mulch
management had a limited impact on cabbage pests and natural enemies. Chapter three results
suggest managing mulch through crimping or allowing it to remain standing is not promising as
a potential habitat management strategy to enhance natural enemies in cabbage. However, the
presence of rye mulch in cabbage plots led to greater abundance of several predators and a
specialist parasitoid. Rye mulch may have potential as a form of structural complexity that can
enhance biological control. Although I observed no impact of mulch species on biological
control of the imported cabbageworm, biological control of other cabbage pests may be impacted
by the presence of mulch.
Additionally, the use of rye mulch for habitat management should be investigated with
other methods of measuring biological control. For instance, molecular gut content analysis
methods can reveal the role of individual predators in managing cabbage pests and identify the
individual predator species providing the greatest pest suppression. Video surveillance methods
also provide insight into predator species feeding in the field and can identify day time and
nocturnal predators of sentinel prey. Further research on habitat management strategies in
cabbage should focus on the natural enemies identified as having the greatest role in cabbage
pest suppression.

109

Although the work in this thesis has provided information on the impact of living and
non-living complexity on arthropods, the mechanisms responsible are still unknown. The
positive response of natural enemies to habitat complexity may be due to the presence of
resources such as detritivore prey, protection from intraguild predators, and suitable
microhabitats. Negative responses of natural enemies to habitat complexity may be due to other
mechanisms such as reduced hunting efficiency and greater intraguild predation. If the
mechanisms responsible for the trends observed in this thesis are identified, habitat management
strategies can be adjusted to maximize the benefits of natural enemy populations in cabbage.
Chapter four of this thesis examined one possible mechanism behind natural enemy
behavior in mulched cabbage. In particular, the role of odor cues from decaying mulch on the
behavior of three natural enemies. The preference for odors from damaged cabbage, rye mulch,
vetch mulch, and combined rye and vetch mulch were observed. The results from chapter four
suggest greater discrimination of specialist natural enemies to damaged cabbage compared to
generalist natural enemies. On the other hand, the findings in chapter four indicate that odors
from decaying mulch or vetch are neither attractive nor repellent to natural enemies and have
limited interfering effects. This indicates a minimal role of olfactory cues in the behavior of
natural enemies to choose mulched habitats. However, the role of odor cues should be examined
in other natural enemies commonly feeding on the imported cabbageworm and other pests before
eliminating the role of olfaction. Based on these results, visual cues are suspected to play a major
role in habitat choice of generalist natural enemies. The impact of vegetational and structural
complexity between crop rows on visual cues for natural enemies has been unexplored to date
and is an avenue for future work.

110

APPENDIX

111

Appendix 1.1. Supplementary data
Supplementary Table S1. Arthropods were categorized according to feeding functional
group, feeding guild, diet breadth, and hunting mode in 2010 and 2011.
Herbivore taxonomic classification
Feeding functional Feeding
Diet breadth
group
guild
Acrididae
Herbivore
Chewing
Generalist
Aphrodinae: Cicadellidae
Herbivore
Sucking
Generalist
Brevicoryne brassicae (Linnaeus)
Herbivore
Sucking
Specialist
Ceramica picta (Harris)
Herbivore
Chewing
Generalist
Chrysomelidae
Herbivore
Chewing
Generalist
Curculionidae
Herbivore
Sucking
Generalist
Delia radicum (Meigen)
Herbivore
Chewing
Specialist
Diabrotica undecimpunctata (Howardi) Herbivore
Chewing
Generalist
Elateridae
Herbivore
Chewing
Generalist
Lygus lineolaris (Palisot de Beauvois)
Herbivore
Sucking
Generalist
Macrosteles fascifrons (Stål)
Herbivore
Sucking
Generalist
Meloidae
Herbivore
Chewing
Generalist
Miridae
Herbivore
Sucking
Generalist
Myzus persicae (Sulzer)
Herbivore
Sucking
Generalist
Phyllotreta cruciferae (Fabricius)
Herbivore
Chewing
Oligophagous
Phyllotreta striolata (Fabricius)
Herbivore
Chewing
Oligophagous
Pieris rapae (Linnaeus)
Herbivore
Chewing
Specialist
Plutella xylostella (Linnaeus)
Herbivore
Chewing
Specialist
Psyllidae
Herbivore
Sucking
Generalist
Spilosoma virginica (Fabricius)
Herbivore
Chewing
Generalist
Thrips tabaci (Lindeman)
Herbivore
Sucking
Generalist
Trichoplusia ni (Hübner)
Herbivore
Chewing
Generalist
Natural enemy taxonomic
classification
Aelothrips sp.
Araneidae
Braconidae
Cantharidae
Carabidae
Chalcididae
Chalcidoidea
Cleridae
Coccinella septempunctata (Linnaeus)
Coleomegilla maculate (DeGeer)
Cotesia rubecula (Marshall)
Diadegma insulare (Cresson)
Dolichopodidae
Harmonia axyridis (Pallas)
Hippodamia convergens (Guérin)

Feeding functional
group
Predator
Predator
Parasitoid
Predator
Predator
Parasitoid
Parasitoid
Predator
Predator
Predator
Parasitoid
Parasitoid
Predator
Predator
Predator

112

Hunting
mode
Active
Sit and wait
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active

Supplementary Table S1 (cont’d)
Natural enemy taxonomic
classification
Ichneumonidae
Lampyridae
Nabidae
Neuroptera
Orius sp.
Phalangidae
Podisus maculiventris (Say)
Propylaea quatuordecimpunctata
(Linnaeus)
Pteromalidae
Salticidae
Staphylinidae
Syrphidae
Tachinidae
Tetragnathidae

Feeding functional
group
Parasitoid
Predator
Predator
Predator
Predator
Predator
Predator
Predator

Hunting
mode
Active
Active
Active
Active
Active
Active
Active
Active

Parasitoid
Predator
Predator
Predator
Parasitoid
Predator

Active
Active
Active
Active
Active
Sit and wait

113

Supplementary Table S2. Mean ± SEM cabbage biomass, leaf area, and number of leaves per plant mid-season in 2010 and 2011.
a
Cabbage leaf area
Number of cabbage leaves
Cabbage biomass (kg)
2 a
a
per plant (cm )
per plant
Main effects
2010
2011
2010
2011
2010
2011
Oat kill date
Early
0.41 ± 0.03a
0.38 ± 0.04
46.44 ± 2.54a
12.58 ± 1.24
17.25 ± 0.55
12.72 ± 0.28
0.34 ± 0.04a
0.35 ± 0.05
39.97 ± 4.07a
11.68 ± 1.73
16.56 ± 1.06
13.16 ± 0.40
Intermediate
Late
0.25 ± 0.04b
0.27 ± 0.07
29.72 ± 4.38b
9.66 ± 2.36
15.44 ± 1.29
12.16 ± 0.56
Herbicide
intensity
Low
0.35 ± 0.03
0.37 ± 0.05
40.35 ± 3.26
12.65 ± 1.70
16.29 ± 0.68
12.85 ± 0.38
0.32 ± 0.04
0.29 ± 0.04
37.06 ± 3.92
9.96 ± 1.16
16.54 ± 0.97
12.50 ± 0.34
High
Significance
Oat kill date
(KD)
Herb.
intensity (HI)
KD x HI
a

---------------------------------------------------------------P-value------------------------------------------------------------<0.01
0.34
<0.01
0.50
0.45
0.31
0.39

0.19

0.36

0.20

0.83

0.50

0.49

0.36

0.52

0.49

0.77

0.63

Different letters within column indicate significant differences within main effects according to LSMeans (α=0.05)

114

Supplementary Table S3. Mean ± SEM number of marketable and non-marketable cabbage heads per
plot in 2010 and 2011.
Marketable
Main effects
Oat kill date
Early
Intermediate
Late
Herbicide intensity
Low
High
Significance
Oat kill date
(KD)
Herb. Intensity
(HI)
KD x HI

2010

a

a

2011

Non-marketable
2010
2011

38.63 ± 0.80a
34.63 ± 2.37a
28.13 ± 3.45b

42.50 ± 2.09
38.88 ± 3.48
36.63 ± 2.18

1.88 ± 0.67a
5.88 ± 2.27a
12.38 ± 3.19b

5.38 ± 1.51
9.63 ± 3.67
12.25 ± 2.02

31.42 ± 2.78
36.17 ± 1.53

38.08 ± 1.99
40.58 ± 2.41

8.50 ± 2.63
4.92 ± 1.62

10.67 ± 2.06
7.50 ± 2.28

------------------------------------P-value-----------------------------------------0.01

0.24

<0.01

0.10

0.08

0.37

0.16

0.22

0.70

0.74

0.78

0.73

a

Different letters within column indicate significant differences within main effects according to
LSMeans (α=0.05)

115

Supplementary Table S4. Mean ± SEM number of non-marketable and
marketable cabbage heads at harvest in 2012.
Main effects
Number of
Number of Nona
a
Marketable Heads
marketable Heads
Mulch species
Bare soil
34.00 ± 4.81a
48.13 ± 7.44a
Hairy vetch
28.63 ± 4.11ab
58.38 ± 4.95ab
Winter rye
16.75 ± 3.04b
73.75 ± 4.30b
Mulch management
No crimping
26.75 ± 4.03
60.58 ± 5.67
Crimped
26.17 ± 3.76
59.58 ± 5.45
Significance
-------------------------P-value---------------Mulch species (S)
0.01
0.02
Mulch management (M)
0.64
0.89
0.25
SxM
0.07
a

Different letters within column indicate significant differences within main
effects according to LSMeans (α=0.05).

116

Oat
kill date

Herbicide
intensity

Replicate
number

Early

Low

1

Late

High

4

Late

High

1

Early

Low

2

Late

Low

2

Early

High

1

Intermediate

Low

4

Intermediate

High

4

Late

Low

1

Intermediate

High

1

Early

Low

3

Early

High

3

Early

Low

4

Late

Low

3

Intermediate

Low

2

Intermediate

Low

1

Intermediate

Low

3

Late

Low

4

Early

High

2

Late

High

2

Intermediate

High

2

Intermediate

High

3

Early

High

4

Late

High

3

A)

0.25

0.75
1.25
1.75
2.25
2.75
3.25
Maximum distance between clusters
Supplementary Figure S1. Community dendogram of plots based on average linkage distance
for herbivores in 2010 (A) and 2011 (B) and natural enemies in 2010 (C) and 2011 (D).

117

Supplementary Figure S1 (cont’d)
Mulch
species

Mulch
management

Replicate
number

Early

Low

1

Early

Low

2

Late

Low

2

Early

High

4

Early

Low

4

Intermediate

Low

1

Late

Low

1

Intermediate

Low

4

Intermediate

Low

3

Late

Low

3

Late

Low

4

Early

High

2

Early

High

3

Intermediate

High

2

Intermediate

High

3

Intermediate

High

4

Late

High

3

Late

High

4

Early

Low

3

Intermediate

Low

2

Early

High

1

Intermediate

High

1

Late

High

1

Late

High

B)

2
0.25

118

0.75
1.25
1.75
2.25
2.75
Maximum distance between clusters

Supplementary Figure S1 (cont’d)
Mulch
species

Mulch
management

Replicate
number

Early

Low

1

Early

Low

2

Intermediate

Low

1

Early

Low

3

Intermediate

Low

4

Early

Low

4

Intermediate

Low

2

Intermediate

Low

3

Early

High

1

Early

High

2

Early

High

3

Intermediate

High

3

Early

High

4

Intermediate

High

2

Late

High

1

Intermediate

High

1

Intermediate

High

4

Late

High

4

Late

Low

1

Late

Low

4

Late

Low

2

Late

Low

3

Late

High

2

Late

High

C)

3
Supplementary Figure S1 (cont’d)
0.25
0.75
1.25
1.75
2.25
Maximum distance between clusters

119

Supplementary Figure S1 (cont’d)
Mulch
species

Mulch
management

Replicate
number

Early

Low

1

Intermediate

Low

2

Early

Low

2

Late

Low

1

Late

Low

2

Late

Low

4

Intermediate

High

4

Early

Low

3

Early

High

3

Early

High

2

Intermediate

High

3

Intermediate

High

2

Late

High

3

Late

High

4

Late

Low

3

Late

High

1

Late

High

2

Early

High

1

Intermediate

High

1

Early

High

4

Early

Low

4

Intermediate

Low

4

Intermediate

Low

1

Intermediate

Low

3

D)

0.25
0.75
1.25
1.75
2.25
2.75
Maximum distance between clusters

120

APPENDIX 1.2
RECORD OF DEPOSITION OF VOUCHER SPECIMENS
The specimens listed below have been deposited in the named museum as samples of those
species or other taxa, which were used in this research. Voucher recognition labels bearing the
voucher number have been attached or included in fluid preserved specimens.

Voucher Number: __2013-04__________
Author and Title of thesis:
Author: Alexandria Bryant
Title: Influence of living and non-living habitat complexity on arthropods in strip-tilled cabbage
Museum(s) where deposited:
Albert J. Cook Arthropod Research Collection, Michigan State University (MSU)
Specimens:
**If lowest taxonomic level is above family, lowest classification used for arthropod is indicated

Supplementary Table S5. Voucher specimens deposited at the Albert J. Cook Arthropod
Research Collection (Michigan State University).
Preservation
Family
Genus-Species
Life Stage Quantity
Anthocoridae
Orius sp.
Adult
1
Pinned
Aphididae
Brevicoryne brassicae
Mixture
multiple
Alcohol
Braconidae
Cotesia rubecula
Adult
12
Pinned
Cantharidae
Adult
1
Pinned
Carabidae
Agonum cupripenne
Adult
1
Pinned
Carabidae
Amara sp.
Adult
1
Pinned
Carabidae
Anisodactylus sanctaecrucis
Adult
1
Pinned
Carabidae
Bembidion quadrimaculatum
Adult
1
Pinned
Carabidae
Bradycellus rupestris
Adult
1
Pinned
Carabidae
Cicindela sp.
Adult
1
Pinned
Carabidae
Colliurus pensylvanica
Adult
1
Pinned
Carabidae
Cyclotrachelus sodalist
Adult
1
Pinned
Carabidae
Harpalus erraticus
Adult
1
Pinned
Carabidae
Poecilus chalcites
Adult
1
Pinned
Carabidae
Pterostichus permundus
Adult
1
Pinned
Carabidae
Scarites quadriceps
Adult
1
Pinned
Carabidae
Stenolophus comma
Adult
1
Pinned
Chrysomelidae
Phyllotreta cruciferae
Adult
1
Pinned
Chrysomelidae
Phyllotreta striolata
Adult
1
Pinned
Chrysopidae
Adult
1
Pinned
Chrysopidae
Larva
1
Alcohol

121

Supplementary Table S5 (cont’d)
Family
Genus-Species
Class Chilopoda
Coccinellidae
Coccinella septempunctata
Coccinellidae
Coleomegilla maculata
Coccinellidae
Harmonia axyridis
Coccinellidae
Hippodamia convergens
Coccinellidae
Hippodamia parenthesis
Coccinellidae
Propylea quatuodecimpunctata
Coenagrionidae
Geocoridae
Ichneumonidae
Diadegma insulare
Lampyridae
Libellulidae
Mantidae
Miridae
Lygus lineolaris
Nabidae
Noctuidae
Melanchra picta
Noctuidae
Trichoplusia ni
Pentatomidae
Podisus maculiventris
Pentatomidae
Podisus maculiventris
Phalangidae
Pieridae
Pieris rapae
Pieridae
Pieris rapae
Plutella
Plutella xylostella
Plutella
Plutella xylostella
Reduviidae
Salticidae
Sphecidae
Staphylinidae
Superfamily
Chalcidoidea
Syrphidae
Tachinidae
Tetragnathidae
Thomisidae
Vespidae

122

Life Stage
Adult
Adult
Adult
Adult
Adult
Adult
Adult
Adult
Adult
Adult
Adult
Adult
Nymph
Adult
Adult
Larva
Larva
Adult
Nymph
Adult
Adult
Larva
Adult
Larva
Adult
Adult
Adult
Adult

Quantity
1
1
1
1
4
1
1
1
1
1
1
1
1
1
2
2
4
1
1
1
2
4
2
4
1
1
1
1

Preservation
Alcohol
Pinned
Pinned
Pinned
Pinned
Pinned
Pinned
Pinned
Pinned
Pinned
Pinned
Pinned
Alcohol
Pinned
Pinned
Alcohol
Alcohol
Pinned
Alcohol
Alcohol
Pinned
Alcohol
Pinned
Alcohol
Pinned
Alcohol
Pinned
Pinned

Adult
Larva
Adult
Adult
Adult
Adult

6
1
1
1
1
1

Pinned
Alcohol
Pinned
Alcohol
Alcohol
Pinned

LITERATURE CITED

123

LITERATURE CITED
Acar, E. B., J. C. Medina, M. L. Lee, and G. M. Booth. 2001. Olfactory behavior of
convergent lady beetles (Coleoptera : Coccinellidae) to alarm pheromone of green peach
aphid (Hemiptera : Aphididae). Canadian Entomologist 133: 389-397.
Adati, T., W. Susila, K. Sumiartha, P. Sudiarta, W. Toriumi, K. Kawazu, and S. Koji. 2011.
Effects of mixed cropping on population densities and parasitism rates of the diamondback
moth, Plutella xylostella (Lepidoptera: Plutellidae). Applied Entomology and Zoology 46:
247-253.
Adedipe, F., and Y. L. Park. 2010. Visual and olfactory preference of Harmonia axyridis
(Coleoptera: Coccinellidae) adults to various companion plants. Journal of Asia-Pacific
Entomology 13: 319-323.
Agelopoulos, N. G., and M. A. Keller. 1994a. Plant natural enemy association in the tritrophic
system Cotesia rubecula - Pieris rapae - Brassicaceae (Cruciferae) 1. Sources of
infochemicals. Journal of Chemical Ecology 20: 1725-1734.
Agelopoulos, N. G., and M. A. Keller. 1994b. Plant natural enemy association in the tritrophic
system Cotesia rubecula - Pieris rapae - Brassicaceae (Cruciferae) 2. Preference of C.
rubecula for landing and searching. Journal of Chemical Ecology 20: 1735-1748.
Agelopoulos, N. G., and M. A. Keller. 1994c. Plant natural enemy association in the tritrophic
system Cotesia rubecula - Pieris rapae - Brassicaceae (Cruciferae) 3. Collection and
identification of plant and fass volatiles. Preference of C. rubecula for landing and searching.
Journal of Chemical Ecology 20: 1955-1967.
Ahuja, I., J. Rohloff, and A. M. Bones. Defence Mechanisms of Brassicaceae: Implications for
Plant-Insect Interactions and Potential for Integrated Pest Management, pp. 623-670,
Sustainable Agriculture, Vol 2.
Alhmedi, A., E. Haubruge, and F. Francis. 2009. Effect of stinging nettle habitats on
aphidophagous predators and parasitoids in wheat and green pea fields with special attention
to the invader Harmonia axyridis Pallas (Coleoptera: Coccinellidae). Entomological Science
12: 349-358.
Altieri, M. A., and L.L. Schmidt. 1987. Mixing broccoli cultivars reduces cabbage aphid
numbers. California Agriculture 41: 24-26.
Altieri, M. A., R.C. Wilson, and L.L. Schmidt. 1985. The effects of living mulches and weed
cover on the dynamics of foliage arthropod and soil arthropod communities in 3 crop systems.
Crop Protection 4: 201-213.

124

Andow, D. A. 1990. Population dynamics of an insect herbivore in simple and diverse habitats.
Ecology 71: 1006-1017.
Andow, D. A. 1991. Vegetational diversity and arthropod population response. Annual Review
of Entomology 36: 561-586.
Andow, D. A. 1992. Population-density of Empoasca fabae (Homoptera, Cicadellidae) in weedy
beans. Journal of Economic Entomology 85: 379-383.
Andow, D. A., A. G. Nicholson, H. C. Wien, and H. R. Willson. 1986. Insect populations on
cabbage grown with living mulches. Environmental Entomology 15: 293-299.
Andow, D. A., and S.J. Risch. 1985. Predation in diversified agroecosystems, relations between
a coccinellid predator Coleomegilla maculata and its food. Journal of Applied Ecology 22:
357-372.
Andow, D.A. 1988. Management of weeds for insect manipulation in agroecosystems, pp. 265301. In M.A. Altieri and M. Liebman (eds.) Weed management in agroecosystems:
ecological approaches. CRC Press, Boca Raton, FL.
Armstrong, G., and R. G. McKinlay. 1997. Vegetation management in organic cabbages and
pitfall catches of carabid beetles. Agriculture Ecosystems and Environment 64: 267-276.
Banks, J. E. 1998. The scale of landscape fragmentation affects herbivore response to vegetation
heterogeneity. Oecologia 117: 239-246.
Batary, P., A. Baldi, G. Szel, A. Podlussany, I. Rozner, and S. Erdos. 2007. Responses of
grassland specialist and generalist beetles to management and landscape complexity.
Diversity and Distributions 13: 196-202.
Bath, B., J. Malgeryd, A. R. Stintzing, and H. Akerhielm. 2006. Surface mulching with red
clover in white cabbage production. Nitrogen uptake, ammonia losses and the residual
fertility effect in ryegrass. Biological Agriculture and Horticulture 23: 287-304.
Bigger, D. S., and W. E. Chaney. 1998. Effects of Iberis umbellata (Brassicaceae) on insect
pests of cabbage and on potential biological control agents. Environmental Entomology 27:
161-167.
Bjorkman, M., P.A. Hamback, R.J. Hopkins, and B. Ramert. 2010. Evaluating the enemies
hypothesis in a clover-cabbage intercrop: effects of generalist and specialist natural enemies
on the turnip root fly (Delia floralis). Agricultural and Forest Entomology 12: 123-132.
Bommarco, R., F. Miranda, H. Bylund, and C. Bjorkman. 2011. Insecticides Suppress
Natural Enemies and Increase Pest Damage in Cabbage. Journal of Economic Entomology
104: 782-791.

125

Bottenberg, H., J. Masiunas, C. Eastman, and D. Eastburn. 1997. Yield and quality
constraints of cabbage planted in rye mulch. Biological Agriculture and Horticulture 14: 323342.
Brainard, D. C., R. R. Bellinder, and A. J. Miller. 2004. Cultivation and interseeding for weed
control in transplanted cabbage. Weed Technology 18: 704-710.
Brainard, D.C., and D.C. Noyes. 2012. Strip-tillage and compost influence carrot quality, yield
and net returns. HortScience 47: 1073-1079.
Brainard, D.C., and R.R. Bellinder. 2004. Weed suppression in a broccoli-winter rye
intercropping system. Weed Science 52: 281-290.
Brandsaeter, L. O., J. Netland, and R. Meadow. 1998. Yields, weeds, pests and soil nitrogen
in a white cabbage living mulch system. Biological Agriculture and Horticulture 16: 291-309.
Broad, S. T., N.A. Schellhorn, S.N. Lisson, N.J. Mendham, R. Corkrey. 2008. Host location
and parasitism of Brevicoryne brassicae in diversified broccoli cropping systems.
Entomologia Experimentalis Et Applicata 129: 166-171.
Broatch, J. S., L.M. Dosdall, J.T. O'Donovan, K.N. Harker, and G.W. Clayton. 2010.
Responses of the specialist biological control agent, Aleochara bilineata, to vegetational
diversity in canola agroecosystems. Biological Control 52: 58-67.
Brust, G. E. 1994. Natural enemies in straw-mulch reduce Colorado potato beetle populations
and damage in potato. Biological Control 4: 163-169.
Burn, A. 2003. Pesticide buffer zones for the protection of wildlife. Pest Management Science
59: 583-590.
Buschke, F. T., and M.T. Seaman. 2011. Functional feeding groups as a taxonomic surrogate
for a grassland arthropod assemblage. African Invertebrates 52: 217-228.
Cai, H. J., Z.S. Li, and M.S. You. 2007. Impact of habitat diversification on arthropod
communities: A study in the fields of Chinese cabbage, Brassica chinensis. Insect Science
14: 241-249.
Carmona, D. M., and D. A. Landis. 1999. Influence of refuge habitats and cover crops on
seasonal activity-density of ground beetles (Coleoptera : Carabidae) in field crops.
Environmental Entomology 28: 1145-1153.
Cohen, J. 1988. Statistical power analysis for the behavioral sciences. Lawrence Earlbaum
Associates, Hillsdale , NJ.
Costamagna, A. C., D. A. Landis, and M. J. Brewer. 2008. The role of natural enemy guilds in
Aphis glycines suppression. Biological Control 45: 368-379.

126

Costello, M. J., and M.A. Altieri. 1995. Abundance, growth-rate and parasitism of Brevicoryne
brassicae and Myzus persicae (Homoptera, Aphididae) on broccoli grow in living mulches.
Agriculture, Ecosystems, and Environment.52: 187-196.
Coudron, T. A., L. C. Mitchell, R. Sun, D. Robertson, N. V. Pham, and H. J. R. Popham.
2012. Dietary composition affects levels of trace elements in the predator Podisus
maculiventris (Say) (Heteroptera: Pentatomidae). Biological Control 61: 141-146.
de la Pena, N. M., A. Butet, Y. Delettre, P. Morant, and F. Burel. 2003. Landscape context
and carabid beetles (Coleoptera : Carabidae) communities of hedgerows in western France.
Agriculture, Ecosystems, and Environment 94: 59-72.
Desneux, N., A. Decourtye, and J. M. Delpuech. 2007. The sublethal effects of pesticides on
beneficial arthropods, pp. 81-106, Annual Review of Entomology.
Dillard, H. R., R. R. Bellinder, and D. A. Shah. 2004. Integrated management of weeds and
diseases in a cabbage cropping system. Crop Protection 23: 163-168.
Eigenbrode, S. D., T. Castagnola, M. B. Roux, and L. Steljes. 1996. Mobility of three
generalist predators is greater on cabbage with glossy leaf wax than on cabbage with a wax
bloom. Entomologia Experimentalis Et Applicata 81: 335-343.
Elton, C. S. 1958. The ecology of invasions by animals and plants. University of Chicago Press,
Chicago, IL.
FAO stats. 2012. Food and Agriculture Organization of the United Nations, Crop Profile:
Cabbage.
Ferran, A., and A. F. G. Dixon. 1993. Foraging behavior of ladybird larvae (Coleoptera,
Coccinellidae). European Journal of Entomology 90: 383-402.
Finch, S., and R.H. Collier. 2000. Host-plant selection by insects - a theory based on
'appropriate/inappropriate landings' by pest insects of cruciferous plants. Entomologia
Experimentalis Et Applicata 96: 91-102.
Finch, S., H. Billiald, and R. H. Collier. 2003. Companion planting - do aromatic plants disrupt
host-plant finding by the cabbage root fly and the onion fly more effectively than nonaromatic plants? Entomologia Experimentalis Et Applicata 109: 183-195.
Finke, D. L., and R. F. Denno. 2006. Spatial refuge from intraguild predation: implications for
prey suppression and trophic cascades. Oecologia 149: 265-275.
Finke, D. L., and R.F. Denno. 2002. Intraguild predation diminished in complex-structured
vegetation: Implications for prey suppression. Ecology 83: 643-652.

127

Franczuk, J., E. Kosterna, and A. Zaniewicz-Bajkowska. 2010. Weed-control effects on
different types of cover-crop mulches. Acta Agriculturae Scandinavica Section B-Soil and
Plant Science 60: 472-479.
Furlong, M. J., and M. P. Zalucki. 2010. Exploiting predators for pest management: the need
for sound ecological assessment. Entomologia Experimentalis Et Applicata 135: 225-236.
Garcia, M. A., and M.A. Altieri. 1992. Explaining differences in flea beetle Phyllotreta
cruciferae Goeze densities in simple and mixed broccoli cropping systems as a function of
individual behavior. Entomologia Experimentalis Et Applicata 62: 201-209.
Giamoustaris, A., and R. Mithen. 1995. The effect of modifying the glucosinolate content of
leaves of oilseed rape (Brassica napus spp. oleifera) on its interaction with the specialist and
generalist pests. Annals of Applied Biology 126: 347-363.
Gingras, D., P. Dutilleul, and G. Boivin. 2003. Effect of plant structure on host finding
capacity of lepidopterous pests of crucifers by two Trichogramma parasitoids. Biological
Control 27: 25-31.
Grevstad, F. S., and B. W. Klepetka. 1992. The influence of plant architecture on the foraging
efficiencies of a suite of ladybird beetles feeding on aphids. Oecologia 92: 399-404.
Grieshop, M.J., B. Werling, K. Buehrer, J. Perrone, R. Isaacs, and D. Landis. 2012. Big
brother is watching: Studying insect predation in the age of digital surveillance. American
Entomologist 58(3): 172-182 (11).
Grzywacz, D., A. Rossbach, A. Rauf, D. A. Russell, R. Srinivasan, and A. M. Shelton.
Current control methods for diamondback moth and other Brassica insect pests and the
prospects for improved management with lepidopteran-resistant Bt vegetable brassicas in
Asia and Africa. Crop Protection 29: 68-79.
Gurevitch, J., and L.V. Hedges. 1999. Statistical issues in ecological meta-analyses. Ecology
80: 1142-1149.
Haddad, N. M., G.M. Crutsinger, K. Gross, J. Haarstad, and D. Tilman. 2011. Plant
diversity and the stability of foodwebs. Ecology Letters 14: 42-46.
Hamilton, R. M., E. B. Dogan, G. B. Schaalje, and G. M. Booth. 1999. Olfactory response of
the lady beetle Hippodamia convergens (Coleoptera : Coccinellidae) to prey related odors,
including a scanning electron microscopy study of the antennal sensilla. Environmental
Entomology 28: 812-822.
Haramoto, E. R., and D. C. Brainard. 2012. Strip Tillage and Oat Cover Crops Increase Soil
Moisture and Influence N Mineralization Patterns in Cabbage. Hortscience 47: 1596-1602.

128

Harmon, J. P., E.E. Hladilek, J.L. Hinton, T.J. Stodola, and D.A. Andow. 2003. Herbivore
response to vegetational diversity: spatial interaction of resources and natural enemies.
Population Ecology 45: 75-81.
Harvey, C. T., and M. D. Eubanks. 2004. Effect of habitat complexity on biological control by
the red imported fire ant (Hymenoptera: Formicidae) in collards. Biological Control 29: 348358.
Harvey, J. A., and T. M. Fortuna. 2012. Chemical and structural effects of invasive plants on
herbivore-parasitoid/predator interactions in native communities. Entomologia
Experimentalis Et Applicata 144: 14-26.
Harvey, J. A., W.H. van der Putten, H. Turin, R. Wagenaar, and T.M. Bezemer. 2008.
Effects of changes in plant species richness and community traits on carabid assemblages and
feeding guilds. Agriculture, Ecosystems, and Environment 127: 100-106.
Haseeb, M., and H. Amano. 2002. Effects of contact, oral and persistent toxicity of selected
pesticides on Cotesia plutellae (Hym., Braconidae), a potential parasitoid of Plutella
xylostella (Lep., Plutellidae). Journal of Applied Entomology-Zeitschrift Fur Angewandte
Entomologie 126: 8-13.
Herlihy, M. V., R. G. Van Driesche, M. R. Abney, J. Brodeur, A. B. Bryant, R. A.
Casagrande, D. A. Delaney, T. E. Elkner, S. J. Fleischer, R. L. Groves, D. S. Gruner, J.
P. Harmon, G. E. Heimpel, K. Hemady, T. P. Kuhar, C. M. Maund, A. M. Shelton, A. J.
Seaman, M. Skinner, R. Weinzierl, K. V. Yeargan, and Z. Szendrei. 2012. Distribution of
Cotesia rubecula (Hymenoptera : Braconidae) and its displacement of Cotesia glomerata in
Eastern North America. Florida Entomologist 95: 461-467.
Hines, R. L., and W.D. Hutchison. 2001. Evaluation of action thresholds and spinosad for
lepidopteran pest management in Minnesota cabbage. Journal of Economic Entomology 94:
190-196.
Hooks, C. R. R., and M.W. Johnson. 2002. Lepidopteran pest populations and crop yields in
row intercropped broccoli. Agriculture and Forest Entomology 4: 117-125.
Hooks, C. R. R., and M.W. Johnson. 2003. Impact of agricultural diversification on the insect
community of cruciferous crops. Crop Protection 22: 223-238.
Hooks, C. R. R., and M.W. Johnson. 2004. Using undersown clovers as living mulches: effects
Hooks, C. R. R., and M.W. Johnson. 2006. Population densities of herbivorous lepidopterans
in diverse cruciferous cropping habitats: Effects of mixed cropping and using a living mulch.
Biocontrol. 51: 485-506.
Horton, D. R., D.A. Broers, R.R. Lewis, D. Granatstein, R.S. Zack, T.R. Unruh, A.R.
Moldenke, and J.J. Brown. 2003. Effects of mowing frequency on densities of natural

129

enemies in three Pacific Northwest pear orchards. Entomologia Experimentalis Et Applicata
106: 135-145.
Hummel, J. D., L.M. Dosdall, G.W. Clayton, K.N. Harker, and J.T. O'Donovan. 2012.
Ground Beetle (Coleoptera: Carabidae) diversity, activity density, and community structure
in a diversified agroecosystem. Environmental Entomology 41: 72-80.
Hutchison, W. D., and E. C. Burkness. 1999. On-farm experience using action thresholds and
SpinTor for the management of Lepidoptera in cabbage: steps to implementation, pp. 35-39.
In Proceedings, International Workshop on Integrated Pest Management of Cole Crops, 2021 May 1999. Universidad de Celaya. Celaya, Mexico.
Ilnicki, R. D., and A. J. Enache. 1992. Subterranean clover living mulch – an alternative
method of weed control. Agriculture Ecosystems and Environment 40: 249-264.
Janmaat, A. F., and J. Myers. 2003. Rapid evolution and the cost of resistance to Bacillus
thuringiensis in greenhouse populations of cabbage loopers, Trichoplusia ni. Proceedings of
the Royal Society B-Biological Sciences 270: 2263-2270.
Janssen, A., M.W. Sabelis, S. Magalhaes, M. Montserrat, and T. Van der Hammen. 2007.
Habitat structure affects intraguild predation. Ecology 88: 2713-2719.
Johnson, J. M., J. A. Hough-Goldstein, and M. J. Vangessel. 2004. Effects of straw mulch on
pest insects, predators, and weeds in watermelons and potatoes. Environmental Entomology
33: 1632-1643.
Johnson, R. M., M. D. Ellis, C. A. Mullin, and M. Frazier. 2010. Pesticides and honey bee
toxicity - USA. Apidologie 41: 312-331.
Jonsson, M. 2009. The impact of floral resource subsidies and landscape composition on
biological control of brassica pests, pp. 208. In P.G. Mason, D.R. Gillespie, and C. Vincent
(eds.), Proceedings of the 3rd International Symposium on Biological Control of Arthropods,
Christchurch, New Zealand, 8-13 February 2009, Christchurch, New Zealand.
Jonsson, M., H. L. Buckley, B. S. Case, S. D. Wratten, R. J. Hale, and R. K. Didham. 2012.
Agricultural intensification drives landscape-context effects on host-parasitoid interactions in
agroecosystems. Journal of Applied Ecology 49: 706-714.
Kaplan, I. 2012. Attracting carnivorous arthropods with plant volatiles: The future of biocontrol
or playing with fire? Biological Control 60: 77-89.
Kawashima, M., and C. Jung. 2010. Effects of sheltered ground habitats on the overwintering
potential of the predacious mite Neoseiulus californicus (Acari: Phytoseiidae) in apple
orchards on mainland Korea. Experimental and Applied Acarology 55: 375-388.

130

Landis, D. A., S.D. Wratten, and G.M. Gurr. 2000. Habitat management to conserve natural
enemies of arthropod pests in agriculture. Annual Review of Entomology 45: 175-201.
Langellotto, G. A., and R.F. Denno. 2004. Responses of invertebrate natural enemies to
complex-structured habitats: a meta-analytical synthesis. Oecologia 139: 1-10.
Larentzaki, E., J. Plate, B. A. Nault, and A. M. Shelton. 2008. Impact of straw mulch on
populations of onion thrips (Thysanoptera: Thripidae) in onion. Journal of Economic
Entomology 101: 1317-1324.
Latheef, M.A., and J.H. Ortiz. 1983. The influence of companion herbs on egg distribution of
the imported cabbageworm, Pieris rapae (Lepidoptera, Pieridae), on collard plants. Canadian
Entomologist 115: 1031-1038.
Laub, C. A., and J. M. Luna. 1992. Winter cover crop suppression practices and natural
enemies of armyworm (Lepidoptera, Noctuidae) in no-till corn. Environmental Entomology
21: 41-49.
Lee, J. C., and G.E. Heimpel. 2005. Impact of flowering buckwheat on Lepidopteran cabbage
pests and their parasitoids at two spatial scales. Biological Control. 34: 290-301.
Lee, J. C., D. A. Andow, and G. E. Heimpel. 2006. Influence of floral resources on sugar
feeding and nutrient dynamics of a parasitoid in the field. Ecological Entomology 31: 470480.
Lee, J. C., and G. E. Heimpel. 2008. Floral resources impact longevity and oviposition rate of a
parasitoid in the field. Journal of Animal Ecology 77: 565-572.
Leibee, G. L., and K. E. Savage. 1992. Evaluations of selected insecticides for control of
diamondback moth and cabbage looper in cabbage in central Florida with observations on
insecticide resistance in the diamondback moth. Florida Entomologist 75: 585-591.
Legrand, A., and P. Barbosa. 2003. Plant morphological complexity impacts foraging
efficiency of adult Coccinella septempunctata L. (Coleoptera : Coccinellidae).
Environmental Entomology 32: 1219-1226.
Letourneau, D. K., I. Armbrecht, B.S. Rivera, J.M. Lerma, E.J. Carmona, M.C. Daza, S.
Escobar, V. Galindo, C. Gutierrez, S.D. Lopez, J.L. Mejia, A.M.A. Rangel, J.H. Rangel,
L. Rivera, C.A. Saavedra, A.M. Torres, and A.R. Trujillo. 2011. Does plant diversity
benefit agroecosystems? A synthetic review. Ecol. Appl. 21: 9-21.
Li, Q., S. D. Eigenbrode, G. R. Stringham, and M. R. Thiagarajah. 2000. Feeding and
growth of Plutella xylostella and Spodoptera eridania on Brassica juncea with varying
glucosinolate concentrations and myrosinase activities. Journal of Chemical Ecology 26:
2401-2419.

131

Liebman, M., and E. Dyck. 1993. Crop rotation and intercropping strategies for weed
management. Journal of Applied Ecology 3: 92–122.
Luna, J. M., and M.L. Staben. 2002. Strip tillage for sweet corn production: Yield and
economic return. HortScience 37: 1040-1044.
Luna, J.M., J.P. Mitchell, and A. Shrestha. 2012. Conservation tillage in organic agriculture:
evolution toward hybrid systems in the Western USA. Renewable Agriculture and Food
Systems 27: 21-30.
Lundgren, J. G., K.A.G. Wyckhuys, and N. Desneux. 2009. Population responses by Orius
insidiosus to vegetational diversity. Biocontrol. 54: 135-142.
Lundgren, J. G., and J.K. Fergen. 2010. The effects of a winter cover crop on Diabrotica
virgifera (Coleoptera: Chrysomelidae) populations and beneficial arthropod communities in
no-till maize. Environmental Entomology. 39: 1816-1828.
Lundgren, J. G., and J. K. Fergen. 2011. Enhancing predation of a subterranean insect pest: A
conservation benefit of winter vegetation in agroecosystems. Applied Soil Ecology 51: 9-16.
Lynch, J. M., and A. M. Zobel. 2002. Kale intercropping for insect control in cabbage crops as
an alternate to insecticides application. Allelopathy Journal 9: 53-57.
Maguire, L.A. 1984. Influence of surrounding plants on densities of Pieris rapae (L.) eggs and
larvae (Lepidoptera, Pieridae) on collards. Environmental Entomology 13: 464-468.
Mangan, F., R. Degregorio, M. Schonbeck, S. Herbert, K. Guillard, R. Hazzard, E.
Sideman, and G. Litchfield. 1995. Cover cropping systems for Brassicas in the
Northeastern United States 2. Weed, insect, and slug incidence. Journal of Sustainable
Agriculture 5: 15-36.
Manrique, V., C. O. Montemayor, R. D. Cave, E. A. Skvarch, and B. W. Smith. Effect of
straw mulch on populations of Microtheca ochroloma (Coleoptera: Chrysomelidae) and
ground predators in turnip Brassica rapa in Florida. Florida Entomologist 93: 407-411.
Marshall, S.A. 2006. Insects their natural history and diversity: With a photographic guide to
insects of North America. Firefly Books Ltd., Buffalo, New York.
Martin, S. G., R.C. Van Acker, and L.F. Friesen. 2001. Critical period of weed control in
spring canola. Weed Science 49: 326-333.
Martinez, J. and G. Guiraud. 1990. A lysimeter study of the effects of a ryegrass catch crop,
during a winter wheat/maize rotation, on nitrate leaching and the following crop. Journal of
Soil Science 41:5–16.

132

Masiunas, J. B., D. M. Eastburn, V. N. Mwaja, and C. E. Eastman. 1997. The impact of
living and cover crop mulch systems on pests and yields of snap beans and cabbage. Journal
of Sustainable Agriculture 9: 61-89.
McGill, B.J., B.J. Enquist, E. Weiher, and M. Westoby. 2006. Rebuilding community ecology
from functional traits. Trends in Ecology and Evolution 21: 178-185.
McNett, B. J., and A.L. Rypstra. 2000. Habitat selection in a large orb-weaving spider:
vegetational complexity determines site selection and distribution. Ecological Entomology
25: 423-432.
Mitchell, E. R., G. G. Ho, and D. Johanowicz. 2000. Management of diamondback moth
(Lepidoptera : plutellidae) in cabbage using collard as a trap crop. Hortscience 35: 875-879.
Mitchell, E. R., G. Y. Hu, and J. S. Okine. 1997. Diamondback moth (Lepidoptera:
Plutellidae) infestation and parasitism by Diadeagma insulare (Hymenoptera:
Ichneumonidae) in collards and adjacent cabbage fields. Florida Entomologist 80: 54-62.
Mochizuki, M. J., A. Rangarajan, R. R. Bellinder, H. M. van Es, and T. Bjorkman. 2008.
Rye mulch management affects short-term indicators of soil quality in the transition to
conservation tillage for cabbage. Hortscience 43: 862-867.
Moreno, C. R., S. A. Lewins, and P. Barbosa. 2010. Influence of relative abundance and
taxonomic identity on the effectiveness of generalist predators as biological control agents.
Biological Control 52: 96-103.
Morse, R. D., and D. L. Seward. 1986. No-tillage production of broccoli and cabbage. Applied
Agricultural Research 1: 96-99.
Muller-Scharer, H., and C.A. Potter. 1991. Cover plants in field growing vegetables; prospects
and limitations. British Crop Prot Conference, Weeds-1991, 1: 599-604.
Mulvaney, M. J., A.J. Price, and C.W. Wood. 2011. Cover crop residue and organic mulches
provide weed control during limited-input no-till collard production. Journal of Sustainable
Agriculture 35: 312-328.
Mumm, R., M. Burow, G. Bukovinszkine'Kiss, E. Kazantzidou, U. Wittstock, M. Dicke,
and J. Gershenzon. 2008. Formation of Simple Nitriles upon Glucosinolate Hydrolysis
Affects Direct and Indirect Defense Against the Specialist Herbivore, Pieris rapae. Journal
of Chemical Ecology 34: 1311-1321.
Mwaja, V. N., J. B. Masiunas, and C. E. Eastman. 1996. Rye (Secale cereale L) and hairy
vetch (Vicia villosa Roth) intercrop management in fresh-market vegetables. Journal of the
American Society for Horticultural Science 121: 586-591.

133

NASS. 2012. Census of Agriculture for Michigan. USDA, National Agricultural Statistics
Service.
Nelson, W. A., B. A. Kahn, and B. W. Roberts. 1991. Screening cover crops for use in
conservation tillage systems for vegetables following spring plowing. Hortscience 26: 860862.
Nicholson, A.G., and H.C. Wien. 1983. Screening of turfgrasses and clovers for use as living
mulches in sweet corn and cabbage. Journal of the American Society for Horticultural
Science 108: 1071-1076.
Norris, R. F., and M. Kogan. 2000. Interactions between weeds, arthropod pests, and their
natural enemies in managed ecosystems. Weed Science 48: 94-158.
Obermaier, E., A. Heisswolf, H.J. Poethke, B. Randlkofer, and T. Meiners. 2008. Plant
architecture and vegetation structure: Two ways for insect herbivores to escape parasitism.
European Journal of Entomology 105: 233-240.
Pekar, S., and J. Benes. 2008. Aged pesticide residues are detrimental to agrobiont spiders
(Araneae). Journal of Applied Entomology 132: 614-622.
Pfiffner, L., H. Luka, C. Schlatter, A. Juen, and M. Traugott. 2009. Impact of wildflower
strips on biological control of cabbage lepidopterans. Agriculture, Ecosystems and
Environment 129: 310-314.
Poletta, G. L., A. Larriera, P. Siroski, M. D. Mudry, and E. Kleinsorge. 2009. Effects of
chemical contaminants on wildlife: identification of biomarkers in a sentinel species, pp.
181-200, Wildlife: Destruction, Conservation and Biodiversity.
Potter, T. L., C.C. Truman, T.C. Strickland, D.D. Bosch, and T.T. Webster. 2008. Herbicide
incorporation by irrigation and tillage impact on runoff loss. Journal of Environmental
Quality 37: 839-847.
Pullaro, T. C., P. C. Marino, D. M. Jackson, H. F. Harrison, and A. P. Keinath. 2006.
Effects of killed cover crop mulch on weeds, weed seeds, and herbivores. Agriculture,
Ecosystems and Environment 115: 97-104.
Randlkofer, B., E. Obermaier, M. Hilker, and T. Meiners. 2010. Vegetation complexity-The
influence of plant species diversity and plant structures on plant chemical complexity and
arthropods. Basic and Applied Ecology 11: 383-395.
Reddy, G. V. P., E. Tabone, and M. T. Smith. 2004. Mediation of host selection and
oviposition behavior in the diamondback moth Plutella xylostella and its predator
Chrysoperla carnea by chemical cues from cole crops. Biological Control 29: 270-277.

134

Richards, O. W. 1940. The biology of the small white butterfly (Pieris rapae), with special
reference to the factors controlling its abundance. Journal of Animal Ecology 9: 243-288.
Riechert, S. E., and L. Bishop. 1990. Prey control by an assemblage of generalist predators –
spiders in garden test systems. Ecology 71: 1441-1450.
Rieux, R., S. Simon, and H. Defrance. 1999. Role of hedgerows and ground cover management
on arthropod populations in pear orchards. Agriculture, Ecosystems and Environment 73:
119-127.
Riley, H., A. K. Loes, S. Hansen, and S. Dragland. 2003. Yield responses and nutrient
utilization with the use of chopped grass and clover material as surface mulches in an organic
vegetable growing system. Biological Agriculture and Horticulture 21: 63-90.
Risch, S.J., D. Andow, and M.A. Altieri. 1983. Agroecosystem diversity and pest control –
data, tentative conclusions, and new research directions. Environmental Entomology 12: 625629.
Root, R.B. 1973. Organization of a plant arthropod association in simple and diverse habitats:
the fauna of collards (Brassica oleracea). Ecological Monographs 43: 95–124.
Rossbach, A., B. Lohr, and S. Vidal. 2005. Generalism versus specialism: Responses of
Diadegma mollipla (Holmgren) and Diadegma semiclausum (Hellen), to the host shift of the
diamondback moth (Plutella xylostella L.) to peas. Journal of Insect Behavior 18: 491-503.
Rypstra, A. L., P.E. Carter, R.A. Balfour, and S.D. Marshall. 1999. Architectural features of
agricultural habitats and their impact on the spider inhabitants. Journal of Arachnology 27:
371-377.
Sant'Ana, J., R. F. P. Da Silva, and J. C. Dickens. 1999. Olfactory reception of conspecific
aggregation pheromone and plant odors by nymphs of the predator, Podisus maculiventris.
Journal of Chemical Ecology 25: 1813-1826.
SAS Institute. 2002-2008. Proc user’s manual, version 9.2. SAS Institute, Cary, NC.
Schellhorn, N. A., and V.L. Sork. 1997. The impact of weed diversity on insect population
dynamics and crop yield in collards, Brassica oleraceae (Brassicaceae). Oecologia 111: 233240.
Schmaedick, M. A., and A. M. Shelton. 2000. Arthropod predators in cabbage (Cruciferae) and
their potential as naturally occurring biological control agents for Pieris rapae (Lepidoptera :
Pieridae). Canadian Entomologist 132: 655-675.
Schmaedick, M. A., and H. M. Shelton. 1999. Experimental evaluation of arthropod predation
on Pieris rapae (Lepidoptera : Pieridae) eggs and larvae in cabbage. Environmental
Entomology 28: 439-444.

135

Schmidt, M. H., U. Thewes, C. Thies, T. Tscharntke. 2004. Aphid suppression by natural
enemies in mulched cereals. Entomologia Experimentalis Et Applicata 113: 87-93.
Schmidt, N. P., M. E. O'Neal, and J. W. Singer. 2007. Alfalfa living mulch advances
biological control of soybean aphid. Environmental Entomology 36: 416-424.
Schmitz, O. J. 2009. Effects of predator functional diversity on grassland ecosystem function.
Ecology 90: 2339-2345.
Schonbeck, M., S. Herbert, R. Degregorio, F. Mangan, K. Guillard, E. Sideman, J. Herbst,
and R. Jaye. 1993. Cover cropping systems for Brassicas in the Northeastern United States
1. Cover crop and vetetable yields, nutrients, and soil conditions. Journal of Sustainable
Agriculture 3: 105-132.
Schroeder, R., and M. Hilker. 2008. The relevance of background odor in resource location by
insects: A behavioral approach. Bioscience 58: 308-316.
Shelton, A. M., J. Plate, and M. Chen. 2008. Advances in control of onion thrips
(Thysanoptera: Thripidae) in cabbage. Journal of Economic Entomology 101: 438-443.
Shrewsbury, P. M., and M.J. Raupp. 2000. Evaluation of components of vegetational texture
for predicting azalea lace bug, Stephanitis pyrioides (Heteroptera : Tingidae), abundance in
managed landscapes. Environmental Entomology 29: 919-926.
Smid, H. M., J.J.A. van Loon, M.A. Posthumus, L.E.M. Vet. 2002. GC-EAG-analysis of
volatiles from Brussels sprouts plants damaged by two species of Pieris caterpillars:
olfactory receptive range of a specialist and a generalist parasitoid wasp species.
Chemoecology 12: 169-176.
Snapp, S. S., S. M. Swinton, R. Labarta, D. Mutch, J. R. Black, R. Leep, J. Nyiraneza, and
K. O'Neil. 2005. Evaluating cover crops for benefits, costs and performance within cropping
system niches. Agronomy Journal 97: 322-332.
Sokal, R.R., and F.J. Rohlf. 1995. Biometry: the principles and practice of statistics in
biological research. Freeman, New York.
Suenaga, H., and T. Hamamura. 2001. Occurrence of carabid beetles (Coleoptera : Carabidae)
in cabbage fields and their possible impact on lepidopteran pests. Applied Entomology and
Zoology 36: 151-160.
Sunderland, K., and F. Samu. 2000. Effects of agricultural diversification on the abundance,
distribution, and pest control potential of spiders: a review. Entomologia Experimentalis Et
Applicata 95: 1-13.

136

Szendrei, Z., and C. Rodriguez-Saona. 2010. A meta-analysis of insect pest behavioral
manipulation with plant volatiles. Entomologia Experimentalis Et Applicata 134: 201-210.
Szendrei, Z., and D. C. Weber. 2009. Response of predators to habitat manipulation in potato
fields. Biological Control 50: 123-128.
Szendrei, Z., M. H. Greenstone, M. E. Payton, and D. C. Weber. 2010. Molecular gut-content
analysis of a predator assemblage reveals the effect of habitat manipulation on biological
control in the field. Basic and Applied Ecology 11: 153-161.
Talekar, N. S., and A. M. Shelton. 1993. Biology, ecology, and management of the
diamondback moth. Annual Review of Entomology 38: 275-301.
Teasdale, J. R. 1998. Cover crops, smother plants, and weed management, pp. 247-270. In J. L.
Hatfield, D.D. Buhler, and B.A. Stewart (Ed.), Integrated Weed and Soil Management. Ann
Arbor Press, Chelsea, MI.
Theunissen, J., C. J. H. Booij, and L. A. P. Lotz. 1995. Effects of intercropping white cabbage
with clovers on pest infestation and yield. Entomologia Experimentalis Et Applicata 74: 7-16.
Trdan, S., D. Znidarcic, M. Kac, and M. Vidrih. 2008. Yield of early white cabbage grown
under mulch and non-mulch conditions with low populations of onion thrips (Thrips tabaci
Lindeman). International Journal of Pest Management 54: 309-318.
Tscharntke, T., R. Bommarco, Y. Clough, T.O. Crist, D. Kleijn, T.A. Rand, J.M.
Tylianakis, S. van Nouhuys, and S. Vidal. 2008. Conservation biological control and
enemy diversity on a landscape scale (Reprinted from Biol. Control, 43: 294-309). Biological
Control 45: 238-253.
Tylianakis, J. M., and C. M. Romo. 2010. Natural enemy diversity and biological control:
Making sense of the context-dependency. Basic and Applied Ecology 11: 657-668.
Tylianakis, J. M., T. Tscharntke, and O.T. Lewis. 2007. Habitat modification alters the
structure of tropical host-parasitoid food webs. Nature. 445: 202-205.
van Poecke, R. M. P., M. Roosjen, L. Pumarino, and M. Dicke. 2003. Attraction of the
specialist parasitoid Cotesia rubecula to Arabidopsis thaliana infested by host or non-host
herbivore species. Entomologia Experimentalis Et Applicata 107: 229-236.
Vidal, S. 1997. Factors influencing the population dynamics of Brevicoryne brassicae in
undersown Brussels sprouts. Biological Agriculture and Horticulture 15: 285-295.
Vuorinen, T., A. M. Nerg, M. A. Ibrahim, G. V. P. Reddy, and J. K. Holopainen. 2004.
Emission of Plutella xylostella-induced compounds from cabbages grown at elevated CO2
and orientation behavior of the natural enemies. Plant Physiology 135: 1984-1992.

137

Warncke, D., J. Dahl, and B. Zandstra, B., 2004. Nutrient recommendations for Michigan
vegetable crops. MSU Extension Bulletin Publication E2934. East Lansing, MI.
Weaver, S. E. 1984. Critical period of weed competition in three vegetable crops in relation to
management practices. Weed Research 24: 317–326.
Weaver, S. E., M.J. Kropff, and R.M.W. Groeneveld. 1992. Use of ecophysiological models
for crop-weed interference: the critical period of weed interference. Weed Science 40: 302–
307.
Wiedenmann, R. N., R. R. Diaz, C. S. Sadof, and R. J. O'Neil. 2004. Ground beetle
(Coleoptera : Carabidae) assemblages in mulched and non-mulched garden plots. Journal of
the Kansas Entomological Society 77: 99-109.
Wilkinson, J. D., K. D. Biever, and D. L. Hostetter. 1983. Resistance of cabbage looper
Trichoplusia ni (Hubner)(Lepidoptera, Noctuidae) to methomyl in Missouri. Journal of the
Kansas Entomological Society 56: 59-61.
Wilson, A. P., J.A. Hough-Goldstein, M.J. Vangessel, and J.D. Pesek. 2004. Effects of
varying weed communities in corn on European corn borer, Ostrinia nubilalis (Hubner)
(Lepidoptera: Crambidae), oviposition, and egg mass predation. Environmental Entomology
33: 320-327.
Winkler, K., F. L. Wackers, A. J. Termorshuizen, and J. C. van Lenteren. Assessing risks
and benefits of floral supplements in conservation biological control. Biocontrol 55: 719-727.
Wold-Burkness, S. J., W. D. Hutchison, J. C. Lee, R. L. Hines, P. C. Bolin, and G. E.
Heimpel. 2005. A long-term survey of parasitoid species composition and parasitism of
Trichoplusia ni (Lepidoptera : Noctuidae), Plutella xylostella (Lepidoptera : Plutellidae), and
Pieris rapae (Lepidoptera : Pieridae) in Minnesota cabbage. Journal of Entomological
Science 40: 211-221.
Woodcock, B. A., and M.S. Heard. 2011. Disentangling the effects of predator hunting mode
and habitat domain on the top-down control of insect herbivores. Journal of Animal Ecology
80: 495-503.
Xu, Q. C., H. L. Xu, F. F. Qin, J. Y. Tan, G. Liu, and S. Fujiyama. 2010. Relay-intercropping
into tomato decreases cabbage pest incidence. Journal of Food Agriculture and Environment
8: 1037-1041.
Xu, Q. C., S. Fujiyama, and H.L. Xu. 2011. Biological pest control by enhancing populations
of natural enemies in organic farming systems. Journal of Food Agriculture and Environment
9: 455-463.

138

Zaniewicz-Bajkowska, A., J. Franczuk, and E. Kosterna. 2009. Direct and Secondary Effects
of Soil Mulching with Straw on Fresh Mass and Number of Weeds, Vegetable Yield. Polish
Journal of Environmental Studies 18: 1185-1190.
Zhao, J. Z., G. S. Ayers, E. J. Grafius, and F. W. Stehr. 1992. Effects of neighboring nectarproducing plants on populations of pest Lepidoptera and their parasitoids in broccoli
plantings. Great Lakes Entomologist 25: 253-258.

139