t ..V:. .
. :-
.. t.

a.

.v

If?

g.

53%

~

4.1”.

a. Etta.

”1.3:: .3.

V

‘ x

fir;
”.31

E“

9.35%

5... .. m.
, m V. y

.1.
z . K! x.
4.5.2:). 3.
.9. n «r . a
. an 5’25

V 4
ham...
.:

7.. {.05.

WW.“

. 1.. ;
flair,
z, I
asiwnfrfi!

.4 ..

fim

m1 , , . :aWerHndu‘hntl‘

'3
I74.

I,

 

 

THESIS
.1
10b 3

$076093”

Date

0-7 639

 

LIBRARY '
Michig" State
University

 

 

 

This is to certify that the

thesis entitled
Quantifying the Roles of Competition and
Niche Separation in Native and Exotic
Coccinellids, and the Changes in the
Co-Iunity in Response to an Exotic Prey

Species presented by

Charles Henry McKeown

has been accepted towards fulfillment
of the requirements for

MS degree in Entomology

 

 

(saw 3%

Major professor

5/ 4/ 03

MS U is an Affirmative Action/Equal Opportunity Institution

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE |

 

APR 2 3 2007

1:0de

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65-p.15

Quantifying the Roles of Competition and Niche Separation
in Native and Exotic Coccinellids, and the Changes in the
Community in Response to an Exotic Prey Species

By

Charles Henry McKeown

A THESIS
Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree

MASTER OF SCIENCE ENTOMOLOGY

2003

ABSTRACT

Quantifying the Roles of Competition and Niche Separation in Native
and Exotic Coccinellids, and the Changes in the Community in
Response to an Exotic Prey Species

By

Charles McKeown

Coccinellids are of enormous importance in natural and agricultural systems.
Their use as biological control agents has moved different species around the globe. In
Michigan two exotic coccinellids are present in the community and their interactions with
each other and the native community was examined to provide a more detailed picture of
community niche separation. Twelve years of data was used to examine the temporal and
spatial dynamics of the exotic species. The dominant members of the community were
then analyzed to elucidate the mechanics of niche separation via temporal, spatial, and
habitat characteristics using spatial and statistical analysis techniques. The third exotic in
the landscape is the recently arrived soybean aphid; the response of the dominant
members of the community to this new food resource was examined for possible
perturbations in the interactions within the coccinellid community. The exotic species
were found not to be in direct competition due to a temporal separation in their niches.
The four dominant coccinellids segregated themselves by microhabitat and displayed
responses to the soybean aphid ranging from none, to a very strong numerical and spatial

response.

Dedicated to my children
Duncan and Devin
and their eccentric but lovable mother,

Kimberly without whom we would all go insane.

iii

ACKNOWLEGEMENTS

I would like to thank my advisor Dr. Stuart Gage for his generous support and
willingness to allow me to wander a corn field for two years. Thanks to Dr. Manuel
Colunga-Garcia for your help, input, patience and expertise.

I would also like to thank my committee: Drs. Ayers, Whalon and Safir.

A special thank you goes to Dr. Lawrence Besaw, for introducing me to entomology as a
science and shepherding me during my undergraduate work.

Partial support for this work was provided by the NSF LTER grant 612560 and the

Michigan Agricultural Experiment Station

iv

Table of Contents

Table of Contents ................................................................................................................ v
List of Tables: ................................................................................................................... vii
List of Figures: ................................................................................................................. viii
Chapter One Introduction ................................................................................................... l
Perspective .................................................................................................................. l

The Importance of Ladybeetles ...................................................................................... 3
Hypothesis: ..................................................................................................................... 4
Chapter 2: Bionomics of the Species Studied ..................................................................... 7
C occinella septempunctata .............................................................................................. 7
Coleomegilla maculata .................................................................................................... 8
Cycloneda munda ............................................................................................................ 8
Harmonia axyridis ........................................................................................................... 9
Aphis glycmes ....................................................... 1]

Chapter Three: Spatial analysis of landscape dominance by two exotic coccinellids,

Harmonia axyridis and C ()ccinella septempunctata ......................................................... 13
Introduction ....................................................................................................................... l 3
Materials and Methods: ..................................................................................................... 16
Study Site: ......................................................................................................................... 16
Results: .............................................................................................................................. 20

Comparison of 1994 and I997 (soybean years): ....................................................... 20

Comparison of 1995 and 1998 (wheat years): .......................................................... 21

Chapter Four: The Microhabitat Preference and Response of the Dominant Members of

the Coccinellid Complex to an Exotic Pest ....................................................................... 29
Introduction: ...................................................................................................................... 29
Materials and Methods ...................................................................................................... 33
Results ............................................................................................................................... 37
Habitat Segregation ....................................................................................................... 37
Aphid Responses: .......................................................................................................... 40
Discussion ............................................................................................ 55
Habitat segregation ....................................................................................................... 55
Aphid response .............................................................................................................. 59
Objectives ..................................................................................................................... 60
Competition: ................................................................................................................. 62
Future directions: .......................................................................................................... 63
References Cited: .............................................................................................................. 65
Appendix 1: Record of voucher specimen deposition ....................................... 73
Appendix 1.1: Voucher specimen inventory .......................................... 74
Appendix 2: Aphid Response Maps ................................................................................. 73
Appendix 3: Height Graphs ........................................................................ 93

vi

List of Tables

Table 1. The results of the interpolation of the trap data. Each Cell represents 1 square

meter with 18897 cells total. .................................................................................... 20
Table 2 Preparation and planting regimens for the study plots in 2001. .......................... 35
Table 3 Plot preparation and planting regimes in 2002 .................................................... 35

Table 4. 2001 ANOVA results and correlation coefficients for the response of the
coccinellids to the habitat type index for 2001 and 2002. If the Pr>F is less than .05
the species showed a significant response to habitat. The correlation coefficient

( R2) indicates the strength of this response. ........................................................... 38

Table 5. 2001 Linear orthogonal contrasts between treatments for each species in 2001.
Contrasts marked with an X are statistically significant using linear orthogonal
contrasts. Significant contrasts show a difference in habitat preferred. ................... 39

Table 6. Linear orthogonal contrasts between treatments for each species in 2002.
Contrasts marked with an X are statistically significant using quadratic linear
contrasts. Significant contrasts show a difference in habitat preferred. ................... 39

Table 7. ANOVA results and correlation coefficients for the response of the coccinellids
to the soybean aphid index for 2001 and 2002. If the Pr>F is less than .05 the species

showed a significant response to the aphid. The correlation coefficient ( R2)
indicates the strength of this response. ..................................................................... 40

Table 8 Results of the linear orthogonal contrast for 2001 aphid scale test. Only the
significant contrasts and models are shown. ............................................................. 42

vii

List of Figures

Figure 1. Flowchart representing the direction of research, the questions asked, and
the analysis performed ...................................................................... 6

Figure 2 Coccinella septempunctata on a soybean plant, the object on the left is a
pupal exuvae ................................................................................. 7

Figure 3 Coleomegilla maculata on corn, this specimen has a larva of a parasitic

wasp emerging from its abdomen ......................................................... 8
Figure 4 Adult Harmonia axyridis ................................................................ 9
Figure 5 Harmonia axyridis larvae ................................................................ 9
Figure 6 Aggregation of Harmonia axyridis .................................................... 1 1
Figure 7 Aphis glycmes feeding on soybean .................................................... 12
Figure 8 Locator map for the Kellogg Biological Station ..................................... 16

Figure 9 Treatments and Plot layout at the Kellogg Biological Station’s Long
Term Ecological Research Site ........................................................... 17

Figure 10 The Berger Parker dominance indexes graphed in 100 dd intervals

for 1994 and 1995 .......................................................................... 22
Figure l l The Berger Parker dominance indexes graphed in 100 dd intervals

for 1997 and 1998 .......................................................................... 23
Figure 12 Results of the inverse distance weighted interpolations of the trap

averages in 400 degree day intervals for 1994 an 1995 ............................... 24
Figure 13 Results of the inverse distance weighted interpolations of the trap

averages in 400 degree day intervals for 1997 and 1998 .............................. 25
Figure 14 The averages per row for the native Coccinellids in 2001 .................... 43, 44
Figure 15 The averages per row for the exotic Coccinellids in 2001 .................... 45, 46

viii

Figure 16 The averages per row for the native C occinellids in 2002 ................... 47, 48

Figure 17 The averages per row for the exotic C occinellids in 2002 .................... 49, 50
Figure 18 H. axyridis and C. septempunctata and aphid scale vs. date for both

years .................................................................................... 51, 52
Figure 19 C. maculata and C. munda and aphid scale vs. date for both years ......... 53, 54
Figure 20 Theoretical models of species habitat preferences and actual results ............ 57

Figure 21 C. maculata dispersion maps with the aphid scale superimposed as
contours for 2001 ....................................................................... 73-77

Figure 22 C. munda dispersion maps with the aphid scale superimposed
as contours for 2001 ................................................................... 78-82

Figure 23 C. septempunctata dispersion maps with the aphid scale superimposed
as contours for 2001 .................................................................. 83-87

Figure 24 H. axyridis dispersion maps with the aphid scale superimposed as
contours for 2001 ....................................................................... 88-92

Chapter One Introduction

Perspective

The importance of lowering the negative impact of agriculture on the environment
has never been higher. The environmental impacts of fertilizers, herbicides, and
insecticides on things such as water quality, have forced agriculture into other means of
nutrient and pest management. Genetically engineered crops have shown promise in this
arena, but the potential collateral effects (such as the potential lethal effects on non-target
organisms, gene flow into surrounding organisms, and negative public perception) have
reduced this impact (Scriber 2001, Ferber 1999). The modern paradigm in agriculture has
been to treat the field as an entity separate from nature and completely controllable by
human inputs. This approach has generated perturbations worldwide in ecosystems. The
core of the problem is that the field cannot be separated from the surrounding ecosystem.
This realization led to the rise of an ecosystem level approach to agriculture that requires
ecologists and growers to know the system they work in, understand the communities and
abiotic factors that interact within the system, and manage crop production within the
confines of nature.

An ecosystem approach to agriculture has inherent advantages (Ramakrishnan
zoooy

0 Reducing energy inputs by reducing the application of synthetic fertilizers and
pesticides

0 Reducing the damage caused by weeds and insects

0 Reducing the amount of disturbance to the surrounding ecosystem

0 Partially restoring a more natural ecosystem energy flow

There are also inherent disadvantages to this approach. The existing paradigm is firmly
entrenched, and changing it is a long and arduous task with extensive research and
documentation of the system required to make good management decisions. This research
is time consuming, expensive, and can be hard to translate into actions for the grower.

A functional ecosystem management approach to landscape also requires a firm
definition of the function desired for any management to be successful in the long term.
There are many functions a system can be managed for such as:

0 Agricultural production

0 Recreation

0 Aesthetic value

0 Hydrological functions such as water retention, groundwater recharge, and in the

case of wetlands, water filtration

0 Preservation of genetic information

0 Preservation of threatened or endangered species
The key to managing for any of these functions is to balance the uses of the landscape
with the potential of future degradation. Sustainable management requires that the
nutrient, water, and thermodynamic inputs and outputs be balanced.

To analyze the interactions of the agricultural field, with its surrounding
ecosystem, the inputs and outputs both floral and faunal, from the surrounding landscape
must be identified and quantified, as well as the outputs from the field to the ecosystem.
The communities present, both beneficial and harmful, must be managed for maximum

positive impact on the crop and minimum negative impact on the ecosystem. To reduce

inputs of pesticides to a minimum, ecological management of landscapes requires the
heavy use of biological control agents to regulate pest populations.

The overarching goal of this research is to gain a better understanding of the
characteristics of the agroecosystems and the communities present therein to develop a
means of assessing the overall health of the system. This allows us to use these
assessments of ecological health to manage for diversity, increase the overall ecosystem
functionality, and to reduce the chemical inputs into agroecosystems to promote long

term sustainability.

The Importance of Ladybeetles

Ladybeetles (Coleoptera, Coccinellidae) are a very important family of insects
both in natural systems and in agricultural systems as biological control agents. They
have been used extensively world wide to control a variety of sofi bodied insects in
various fruit and row crop systems. The first documented use of a coccinellid as a
biological control agent was the importation of an Australian species into California to
control cottony cushion scale in 1888 (Gordon 1985). C occinellids with the exception of
a few phytophagous species, notably the mexican bean beetle, which can be a serious pest
(Gordon 1985), and a few fungivores in the genus Psyloibora, are voracious predators,
most as larvae as well as adults. An adult can eat roughly 400 aphids before laying eggs
and a single adult may eat 5000 aphids in its lifetime. The larvae can consume up to 350
aphids before pupating (Lyon 2002). This appetite has led to repeated introductions of
coccinellids in North America; these introductions have been beneficial to growers.
However, there is increasing concern about the impacts of exotic species on endemic

communities (Howarth 1991, Miller and Applet 1993, Thomas and Willis 1998).

There are roughly 5,000 coccinellid species worldwide and about 400 in North
America (Gordon 1985). The coccinellid community is a member of a larger
aphidophagous guild that includes lacewings, dragonflies, and others. The size and
distribution of this community is, of course, dependent on the size and distribution of its
prey species. Most coccinellids are generalist aphid predators, capable of feeding on
aphids that use a wide variety of host plants. In times of especially limiting resources,
they will cannibalize the larvae and adults of their own community, even their own
species (Cottrell and Yeargan, 1998‘). C occinellids are very mobile and continuously
move through habitat in search of prey; up to 50% of the adult coccinellids in a field may
move to a new field each day and be replaced by beetles from the surrounding fields
(Frazer 1988).

C occinellids are an excellent group to study because they are easily sampled and most
are readily identified in the field. The fact that they are present in most habitats and feed
on a large variety of pests makes them a good possible indicator of the health of an
ecosystem and the system’s trophic structure. They are a diverse group, and the number
of species of ladybeetles present in a landscape can, with the proper ecological
knowledge, be used as an indicator of the landscape diversity and the level of
fragmentation. Knowledge of the structure and functional characteristics of the
coccinellid community could be an important component of the ecological knowledge

necessary to assess the health of an agroecosystem.

Hypothesis:

Although there is a wealth of knowledge on ladybeetles, the community as a

whole has not been studied at a landscape level spatial scale enough to develop reliable

indicators of ecosystem function using coccinellid community structure. The recent
addition of both an exotic prey species and two exotic Coccinellids further underscores
the need for specific knowledge of how this community functions in an agroecosystem.
The specific hypotheses tested in this thesis were designed to provide a more complete
picture of the ecology of Coccinellids as a group at a large spatial scale. The specific
hypotheses were:
0 That the separation in the niches of the two exotic coccinellids in this system is
temporal.
0 The community will avoid direct competition by segregating species according to
habitat type and temporal variations.
0 The community of coccinellids will respond to the presence of an exotic pest
species with changes in their temporal and spatial dynamics.

The direction of the research and the location in this thesis are shown in figure 1.

Research Direction

 

 

/ What is the mechanism that allows two dominant
generalist
predators to coexist in the same habitats?
\ Chapter 3: Temporal Niche Separation Between

 

 

 

K

 

K

 

How does the Coccinellid community \
minimize
Intraspecific competition?
Chapter 4: Community Microhabitat
Segregation j

 

 

 

K What effects on this system does

the arrival of the soybean aphid have?
Chapter 4: Analysis of Exotic Prey
\ Species .

 

 

 

[ Chapter 5: Summary ]

 

W

 

J

Figure 1. Flowchart representing the direction of research, the questions asked, and the
analysis performed. Images in this thesis are presented in color.

 

Chapter 2: Bionomics of the Species Studied

Coccinella septempunctata

Coccinella septempunctata (L) is a large red ladybird beetle easily recognized by
the seven black spots on the elytra. It is a palearctic species that was introduced
repeatedly in the United States from 1956 to 1971 (Angelet and Jaques, I975). The
intentional introductions are thought to all have failed to establish a population. In 1973
(Angelet et a1) a population was found in
New Jersey, this population is thought to
by the result of an accidental
introduction. The species has since

spread and successfully established

 

populations in most of the Eastern

Figure 2 C occinella septempunctata on a soybean
plant, the object on the left is a pupal exuvae. United States. The genus Coccinella is
Photo: C. McKeown. Images in this thesis are
presented in color.

 

primarily aphid predators, feeding on a

 

wide variety of aphids (Gordon 1985). This generalist nature has allowed the rapid spread
of C. septempunctata. Coccinella septempunctata is commonly found in row crops, old
field habitat, and some perennial crops (Maredia et al 1992). The adults overwinter in
small groups under the leaf litter and emerge in spring to begin feeding (Gordon, 1985).
Mating takes place immediately after emergence, and the eggs are laid in late spring. The
larvae are also generalist aphidophagous predators, feeding on the same types of prey as
the adults (Gordon, 1985). The population of C. septempunctata peaks in mid summer,

then slowly declines through late summer and fall (LTER unpublished data).

Coleomegilla maculata

Coleomegilla maculata Degeer is a medium sized elongate red to yellow ladybird,
common throughout the Eastern United States. It is primarily an aphid predator, but up to
50% of its diet can consist of pollen (Gordon,
1985). It overwinters under leaf litter in
hedgerows near field edges, generally close to
dandelion or other early spring pollen sources
(Colunga-Garcia, 1996). It is more common in

corn and other row crops than in less disturbed

 

habitat. In Michigan it is the most common

Figure 3 C oleomegilla maculata on corn.

this specimen has the larvae of a coccinellid in com. Mating occurs in the fall,
parasitic wasp emerging from its
abdomen- ”‘0‘0‘ 0 MCKeown- Images and the females overwinter mated. In the late

in this thesis are presented in color.

 

spring, after the establishment of prey
populations, the eggs are laid in close proximity to aphid colonies (Maredia et al l992b).

This species can have up to five generations per year (Gordon, 1985).

Cycloneda munda

Cycloneda munda Say is a small ladybird with yellow to orange elytra lacking
any spots. It is primarily an aphid predator, but it has been known to attack other sofi
bodied insects (Gordon, 1985). It has been shown to be primarily an arboreal species
(Maredia et al, 1992) and is common in woodlot and old field habitat. In late summer it
has been shown to move from deciduous habitats to alfalfa and horseweed. It is also

found in com in low numbers (Maredia et al, l992b).

Harmonia axyridis

Harmonia axyridis Pallas is one of the newest members of the coccinellid
community in the United States. As a newer species here, its life history and
characteristics in the United States are not familiar. This summary review of H. axyridis
is longer than the other community members and more thorough to give a larger amount
of background information.

Harmonia axyridis is a large red to pale yellow beetle with varying numbers of
spots (Lamana and Miller 1995). Eggs are laid midsummer and hatch in three to five days
(depending on ambient temperature), and the larvae actively start preying on anything
and everything they can catch. The larval stage lasts from 12 to 14 days and pupation
from five to six days. The next generation of adults emerges at around 1000 degree-days
and feed and mate continuously

until just before entering

. " \137'?3
;_. 1.5 ‘ )d'ltthtg- : ..-.
9 .

diapause. The females overwinter

 

Figure 4 Adult
Harmonia axyridis. Figure 5 Harmonia axyridis

Photo: J.Ogrodnick. larvae. Photo: M. H. Rhoades
Images in this thesis Images in this thesis are .
are ”resented in color. presented in color. known to surv1ve for up to three

unmated. Adults have been

 

 

 

years under optimal conditions (IPM of Alaska, 2003).

Harmonia axyridis ’ establishment in North America was either by intentional
introduction or accidentally via commerce (Day et al. 1994). Independent established
populations were first documented in 1988 (Chapin and Brou 1991). There is very little
documentation of the dispersion and establishment of H. axyridis in the United States, so

the timeline and rate of spread are not well known. Separate accidental or intentional

introductions have established H. arvridis in the Pacific Northwest and also in California
(Lamana and Miller 1998).

This species has now become a nuisance pest in Michigan due to its drive to
aggregate and over winter in buildings. Aggregations of several thousand individuals
over wintering in homes and businesses are not uncommon (Tedders, 1994). It feeds on
most soft bodied insects including popular aphid, soybean aphid, mealwonn, spruce
aphid, coccinellid larvae, as well as pollen, nectar, and fruit as alternate energy sources
(Lamana and Miller 1998).

Harmonia axyridis has effectively exploited a wider array of habitats than most
other coccinellids. When the introductions were being made, it was thought to be a
primarily arboreal species (McClure, 1987) but has since proven adept at exploiting
various habitats such as:

0 Urban and rural development where it can find prey on a large number of

horticultural and ornamental plants (Lamana, and Miller 1998).
0 Row crop agriculture where it is common in soybean, wheat, alfalfa, and corn
(Colunga-Garcia and Gage 1998).

This breadth of suitable habitat shows an expanded ability to find, identify, and
successfully capture various prey species (Obata, 1986). This ability includes the ability
to determine suitable habitat, find a variety of prey, choose appropriate oviposition sites,
and find non-prey food sources when prey is scarce (Obata, 1996).

The overwintering and spring dispersal behavior also gives H. axyridis a
rapid dispersal mechanism. Harmonia axyridis, at around 1400 degree days (personal

observations), leaves its summer habitat and flies north. When an individual encounters a

10

substantial vertical surface i.e. a building, bluff, or large tree, then it searches for a crack
or crevice and moves insides to overwinter (Nalepa et al, 2000). This is a gregarious
phase, and if populations are
high, these aggregations can
become quite large (Figure 6).
The range expansion of H.
axyridis has been augmented

greatly by this tendency. The

 

aggregation can amass in any

Figure 6 Aggregation of Harmonia axyridis. Photo: R.
Mizell. Images in this thesis are presented in color.

 

structure and recreational
vehicles, horse and utility trailers, aircraft, and shipping containers (Pecan South, 1993)

when moved in the spring the dispersal can be rapid and far-reaching.

Aphis glycines

Aphis glycines Matsumura is a recent arrival in the United States. It is a small
yellow to green aphid with dark comicles. It was first detected in the Midwest in 2000
and has since been documented from Minnesota to New York. In the summer it feeds
primarily on soybean where it could become a major pest . The life cycle of the soybean
aphid is complex as is typical for most aphids, according to Takahashi et a1 (1993). It
overwinters on buckthom, a native weed commonly found in hedgerows, and along the
edges of woodlots. In midsummer it moves into soybean and feeds on phloem. Only
female that reproduce parthnogenically are present in the summer, primarily in the
Wingless form. The females reproduce rapidly and mature quickly (seven days), allowing

for up to 15 generations annually in soybean. A population doubling can occur in as little

ll

as three days on quality soybean. If crowding is detected or the quality of the soybean
declined, winged adults appear and reproduce, transferring live nymphs to new sites. In
the fall in the soybean dry
down phase, winged
males and females are
produced, and they
actively seek out
buckthom to mate and
overwinter on. They

typically go through three

 

or four generations on

Figure 7 Aphis glycines feeding on soybean. Photo: Peter
Desborough. Images in this thesis are presented in color.

buckthom before moving

 

into soybean in the spring (Ostlie, 2002)

It has been demonstrated that A. glycmes is capable of transmitting the soybean
mosaic virus, and it is thought that it can transmit other soybean pathogens although no
significant crop damage has yet been recorded. The potential economic impact of this
species has been estimated at an annual loss of 435 million bushels or $2.2 billion in the

North Central Region alone (Ostile, 2002).

Chapter Three: Spatial analysis of landscape dominance
by two exotic coccinellids, Harmonia axyridis and
Coccinella septempunctata

Introduction

The role of interspecific competition in ecology has been the subject of a long and
continuing debate. Hypotheses on the subject range from placing it as the primary driver
of community ecology to disregarding it as a force almost entirely (Connell, 1980).
Competition, as defined by Ricklefs (I997), is “any use or defense of a resource by one
individual or population that reduces the availability of that resource to other individuals
or populations.” The resource can be a food source, habitat, mate, or anything else that an
individual needs to complete its life cycle and successfully reproduce. Schroener (I997)
classified the mechanisms of competition into these categories:

0 Consumptive competition, based on the use of some renewable resource

0 Preemptive competition, based on the occupation of open space

0 Overgrth competition, which occurs when one individual grows upon or

over another, thereby depriving the second of some resource

0 Chemical competition, by production of a toxin that acts at a distance after

diffusing through the environment

0 Territorial competition (defense of space)

0 Encounter competition, which involves transient interaction over a resource

that may result in physical harm, loss of time or energy, or theft of food
Charles Darwin thought that competition should be most intense among closely related

species. “As species of the same genus have usually, though by no means invariably,

some similarity in habits and constitution, and always in structure, the struggle will
generally be more severe between species of the same genus, when they come into
competition with each other, than between species of distinct genera” (Darwin, 1926). As
biological control efforts and accidental introductions establish species in new places, the
potential for competitive exclusion of native species and conflict between different
introduced species increases.

C occinellid beetles have been introduced repeatedly in the United States as
biological control agents. As adults and larvae, they are voracious predators of soft
bodied organisms such as mealy worms, aphids, and mites. While these predators are
beneficial biological control agents in forest and agricultural settings, there is a growing
body of evidence indicating negative impact on native species (Howarth 1991, Miller and
Applet 1993, Thomas and Willis 1998). The evidence indicates that some imported
predators are driving native species to local extinction due to competitive exclusion
(Wheeler and Hobecke I995, Elliot et al. 1996, Colunga-Garcia and Gage 1998).

Two introduced species Coccinella septempunctata (L.) and Harmonia axyridis
Pallas) are commonly found together in the same habitats (Colunga-Garcia and Gage
1998). As generalist predators with a wide habitat range and a high degree of behavioral
plasticity, the two species share many biological traits. Both species were introduced
repeatedly into the United States as biological control agents, C. septempunctata in the
1970’s (Gordon 1985) and H. axyridis in the 1980’s (Day et al 1994). While these two
species are effective biological control agents, their effects on endemic aphidophagous

predator populations is unclear.

l4

Once H. axyridis was established in the United States it became the dominant
coccinellid in the north central region (Upper Great Lakes). Within two years of its
detection at the Kellogg Biological Station (KBS) in Hickory Comers Michigan in July
of 1994, it quickly became the most abundant coccinellid in the Long Term Ecological
Research Site (LTER) (Colunga-Garcia and Gage 1998). Initially thought to be primarily
an arboreal species, H. art-midis quickly spread to all of the LTER agricultural and old
field habitats, exhibiting either a generalist nature or high behavioral plasticity and
adaptability (C olunga-Garcia and Gage 1998).

Having arrived more than 30 years ago C. septempunctata was well established in
the United States before the arrival of H. axyridis. Native to Western Europe it was
repeatedly introduced as a bio-control agent in North America beginning in the 1970’s
(Gordon 1985). It has proven able to adapt to and exploit a number of habitats and is
quite successful in agricultural systems (Maredia et al 1992, Colunga-Garcia and Gage
1998).

C. septempunctata and H. axyridis are aphidophagous predators and both are now
common throughout Eastern North America. They are found in the same habitats and
feed on the same prey. This niche overlap would seem to result in direct competition
between the adults of the two species. Examining the population trends of C.
septempunctata before and after the arrival of H. axyridis revealed neither significant
downward population trend nor a shift to marginal habitat or prey (Colunga-Garcia and
Gage 1998). The hypothesis of this study was that the lack of observable competition
between the adults of these two species was the result of a temporal separation between

their niches.

15

Materials and Methods:

The arrival of H. axyridis had little impact on the numbers of C. septempunctata
present in the LTER (Colunga—Garcia and Gage 1998). The KBS LTER was designed to
represent the common landscapes in the North Central Region of The United States. One
of the principal components of the LTER research is the long term dispersal and
composition of beneficial arthropod predators. The data gathered at the KBS LTER spans
13 years in the same experimental plots. Using this large contiguous dataset allows
analysis of patterns of species composition, dispersal, and the specifics of niche
separation of these two exotic species.

Study Site:

The site of this study was the Long Term Ecological Research Station (LTER) at
Kellogg Biological Station (KBS) in Hickory Comets, Michigan. Southwest Michigan is
', characterized by a diverse mix of habitat
,' including agricultural, pasture, oldfield,

- orchard, and woodland landscapes

‘ x M. (Burbank et al 1992). Established in 1980
~ by the National Science Foundation, the
KBS LTER was designed to monitor the

long term effects of various cropping

 

practices on the environment. One of the

Figure 8. Locator map for the Kellogg
Biological Station. Images in this thesis are
presented in color.

primary components of this research has

been the long term monitoring of

 

arthropod predators. The resultant data from this monitoring was used for part of this

research and analysis. The coccinellid community structure including an exotic
established prior to monitoring is well documented. The dataset covers 13 years both
before and after the arrival a second exotic coccinellid, Harmonia axyridis and an exotic
pest, Aphis glycines in Southwest Michigan. This has allowed monitoring of the invasion
and establishment of two exotic species and their interactions with a previously
established exotic as well as the endemic coccinellids. The main site treatments on the

LTER were planted in wheat and soybean during the years analyzed.

LTER Treatments and The KBS LTER
Trap Locations consists of six repetitions of

seven agricultural treatments
- High input- Cnv till
[1:] High input- No till
[:3 Low input- Cnv till
- Perennial: Alfalfa
[:1 Perennial: Poplar
W Second. Succession
- Zero input- Cnv till

I'll treatments are subject to an
I... annual rotation of corn,

(Figure 9), including four
tilled treatments, sucessional

treatments, and two

  
 

perennial (alfalfa and poplar)

treatments. The tilled

 

soybean, and wheat. The

Figure 9. Treatments and Plot Layout at the Kellogg LTER plots are
Biological Station's Long Term Ecological Research Site.

Images in this thesis are presented in color. .
approx1mately one hectare

each. The plots contain five geo—positioned sampling stations where coccinellid adults
were sampled using double-sided, yellow cardboard sticky traps (22.5 x 14.0 cm)

attached 1.2 m above the ground to a metal stand as described by Maredia et al. (19923).

These locations have been sampled weekly from May to August every year since 1987.
C occinellid adults caught on traps were identified, counted, recorded, and removed every
week. The sticky cardboards were replaced every second week.

Weekly weather data collected by a permanent weather station on the LTER site
was used to determine the degree day progression for each growing season. A 50° (F)
base temperature was used to transform the data set from weekly data to degree day
intervals. This transformation is necessary to allow direct comparison of data across years
at the same points in coccinellid and crop development. The intervals analyzed spatially
were 400, 800, and 1200 degree days. These intervals represent early, mid, and late
growing season times in Michigan. The interval population numbers represent the
average trap catch per location in the field.

The data for each species was then mapped using Arc GIS 3.2. Each trap average
was interpolated onto the 12 nearest raster cells using the inverse distance weighted
method resulting in raster layers with one square meter resolution. The maps were
superimposed on a high resolution aerial black and white photograph to illustrate the
surrounding habitat. Each cell in the maps of C. septempunctata was then subtracted from
each cell of H. axyridis resulting in a one meter resolution raster map (18897 cells total)
of relative species abundance in the LTER landscape. Positive numbers indicate a
preponderance of H. axyridis negative C. septempunctata. In 1994 and 1997 soybean
was planted in the four LTER row crop treatments, and in 1995 and 1998 wheat was the
main site cultivar. Corn was planted on 1996 and 1999, and these years were not
analyzed due to the very low abundance of C. septempunctata in corn as it

overwhelmingly prefers soybean or wheat (Maredia et a1 l992b). Harmonia axyridis is

18

found in corn but not at nearly the numbers in soybean or wheat. The Raster maps were
then analyzed using Arc View 3.1 ’3 map query functions to assess the total area covered

by each species at the three intervals, percent neutral coverage was also calculated.

The relative dominance (d) of the species over the rest of the coccinellid

community was estimated via the Berger-Parker equation d = N m / N where Nmax is

it)!

the number of insects of the species being examined, and N is the number of all insects

ll)!

of all species measured in the sample (Magurran 1988). The entire dataset of 14

coccinellids was used to establish N so the dominance indexes generated are not

simply H. axyridis vs. C. septempunctata, but vs. the entire coccinellid community
present in the LTER. This index produces a measure of proportional abundance while
having low sensitivity to sample size and it is independent of the number of species
(Southwood 1978). Differential attraction between the species to the traps was the largest
possible source of error, the attraction and color response was tested during the
preliminary LTER design (LTER unpublished data). The relative dominance of C.
septempunctata and H. axyridis was calculated in 100 degree day intervals ending at
1200 degree days. The exception being 1995 where monitoring was stopped at 1100
degree days because the traps had to be removed early due to logistical considerations.
The dataset was tested to see if it met the ANOVA assumptions and then the
populations of the two species were then regressed using the model; y=mx+b, y= H.
axyridis population mean, m = slope or the best fit line, x = C. septempunctata
population mean and b = the x intercept. Least squares linear regression was the best fit,

with alpha = .05.

I9

Results:

Comparison of 1994 and 1997 (soybean years):

The actual percent of the LTER area dominated by either species as well as the
area that neither had preponderance is shown in table 1. In 1994 before the arrival of H.
aryridis, C. septempunctata coverage peaked mid season and declined later. The same
pattern holds true for 1997, the other year where soybean was the main LTER cultivar.
The establishment of H. axyridis in late 1994 represented a small area of the total area
(1.7%). In stark contrast, by the next soybean year in 1997 the proportion of the
landscape dominated in the late season by H. axyridis had grown to 72%, a 42 fold
increase in the same crop in 2 years. Even with this substantial increase in H. axyridis, C.
septempunctata maintained its dominance of the landscape in the midseason dominating

62% of the landscape at the 800 dd interval in 1997.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Year Cells H. Percent H. Cells C. Percent C. Neutral Percent
Degree axyridis axyridis septempunctata septempunctata Cells Neutral
Day Coverage Coverage Cells
interval

1 994

400 0 0 10742 57 8255 44

800 0 0 18505 98 492 2.6
1200 322 1.7 12831 68 5844 31
1995

400 1227 6.5 6658 35 11012 58

800 100 0.5 16972 89 1925 10
1200 11023 58 387 2.0 7407 39

1 997

400 1 1666 61 2669 14 4562 24

800 2166 1 1 1 1850 62 4981 26
1200 13716 72 347 1.8 4934 26

1 998

400 2408 13 15286 80 1303 6.9

800 1661 8.8 16795 89 541 2.9
1200 18575 98 51 0.27 371 1.96

 

 

 

 

 

 

 

Table l. The results of the interpolation of the trap data. Each Cell represents 1 square meter with
18897 cells total.

 

Comparison of 1995 and 1998 (wheat years):

In wheat the early season was dominated by C. septempunctata with 35%
coverage in 1995 and 80 % in 1998. The midseason was also clearly dominated by C.
septempunctata with 89 % coverage in both years. However, the late season was
characterized by a resurgence of H. axyridis in both years with 58% in 1995 and 98% in
1998. In 1995 H. avg-midis increased its coverage from 0.5 % at 800 degree days to 58%
40 degree days later. The same pattern is present in 1998 where it increased from 8.8% to
98 % habitat coverage over the same interval. The landscape in 1995 was also
characterized by comparatively low numbers of both coccinellids and a larger proportion
of neutral area across all three time intervals.

The relative dominance index shows clearly that the two species discussed here
clearly dominate the system. Of the fourteen species of coccinellid only Coleomegilla
maculata approached the population of H. axyridis and C. septempunctata. The spatial
representation of species flux, figures 12 and 13, clearly shows that H. axyridis, once
established in the LTER, is more abundant than C. septempunctata in the late season
(1200 degree days). C. septempunctata is clearly more prevalent mid season (800 degree
days). There is no clear pattern of early season (400 degree days) dominance. The pattern

of relative

21

    

 

o . ——a-—- »

~ 300 I 400 . 500 [600 l 700 l 800 I 900 ‘ 1000'1100] 120;!

 

 

Degree Day Interval

Berger Parker Dominance Index

 

 

Degree Day Interval

+ C. septempunctata
_.__ H. axyridis

Figure 10. The Berger Parker Dominance Index at 100 degree day intervals for 1994 and 1995. H.
axyridis makes its first appearance in 1994 in the late season and a year later begins to dominate C.
septempunctata at the 900 degree day interval. Images in this thesis are presented in color.

22

0.8

0.6

0.4

0.2

 

 

 

 

300 I 400 I 500 I 600 I 700 I 800 I 900 1000' 1100I 1200

Degree Day Interval

 

    

Berger Parker Dominance Index

 

 

 

300 I 400 I 500 I 600 700 I 800 I 900 I 1000I 11OOI 1200I

 

Degree Day Interval

+ C. septempunctata
___.,__ H. axyridis

Figure 11. The Berger Parker Dominance Index at 100 degree day intervals for 1997 and 1998. C.
septempunctata show slight dominance in the early season both years but the crossover to H axyridis
dominance between 800 and 900 degree days is consistent. Images in this thesis are presented in
color.

 

 

. s 4

1,1: ‘ _ ‘ 400 degree days 800 degree days 1200 degree days
at: 199

“3.4 . - ' . ' '

 

 

 

  

   

400 degree days 800 degree days 1200 degree days

Figure 12. Results of the inverse distance weighted interpolations of the trap averages in 400 degree
day intervals for 1994 and 1995. The maps show C. septempunctata dominance in blue, and H.
axyridis dominance in red according t the legend above. Areas where neither species is dominant are
transparent revealing the background photograph. The spatial resolution of the interpolations is 1
square meter. Images in this thesis are presented in color.

dominance is consistent and repeats annually, with C. septempunctata dominating the
mid season of each year and H. axyridis dominating the late season. Also apparent is the
speed with which H. axyridis came to dominate the landscape alter its first appearance in
late 1994 (tablel). The regression analysis of the two populations shows that very little of
the variation in the population level of either species is due to the population level of the
other, R2=0.0009 alpha = .05. Also a small amount of habitat preference separates the two

species. H. axyridis was common in the poplar treatments while C. septempunctata

24

 

III-
than.
33.3.5:

_I
nud905

I.

 

EBB
an» -

 

 

 

 

400 degree days 800 degree days 1200 degree days

Figure 13. Results of the inverse distance weighted interpolations of the trap averages in 400 degree
day intervals for 1997 and 1998. The maps show C. septempunctata dominance in blue, and H.
axyridis dominance in red according t the legend above. Areas where neither species is dominant are
transparent revealing the background photograph. The spatial resolution of the interpolations is 1
square meter. Images in this thesis are presented in color.

was uncommon. Coccinella septempunctata shows clear preference for the sucessional
and agricultural treatments over the poplar habitats. The pattern of community dominance
repeats and is consistent for the mid and late seasons regardless of the total abundance of
the two species.

There is some spatial segregation based on the LTER treatment regime, the
notable exception being the lack of C. septempunctata in the poplar treatments that H.

axyridis uses.

25

Discussion:

The abundance and dispersion patterns of C. septempunctata did not change due
to the arrival of H. axyridis in 1994 (Colunga and Gage 1998). This indicates that the
long term fluctuations in the C. septempunctata population are due to some other factor.
The long term population trend of C. septempunctata appears to exhibit a five to six year
population cycle independent of the rest of the coccinellid community (LTER
unpublished data). The pattern of mid season dominance followed by a late season
decline in numbers as well as a contraction in dispersal repeated in the presence of H.
axyridis.

The 400 degree day interval shows no clear pattern of dominance by either
species. This could be the result of the LTER treatment. In wheat years C.
septempunctata covered most of the habitat, while the early season soybean seemed to
favor H. axyridis, but only one soybean year was available for analysis afier H. axyridis'
establishment. To determine if early season dominance is correlated with the LTER
cultivar more soybean data needs to be analyzed.

Other interactions between the two species allow some additional interpretation.
The point where the two dominance indexes crossed in 1995, 1997, and 1998 is
remarkably consistent. Over the three years, it occurred within a 100 dd window,
indicating that the population cycles of both species may be highly regulated. This
regulation could be genetic, behavioral, related to generation times, or due to some other
unknown environmental factor. The level of stochasticity in the dominance indexes in
1997 and 1998 indicate that the temporal segregation between the two species may be

declining due to H. axyridis having become established. The rest of the system may be

26

adjusting to its presence, lowering its dominance. This is also evidenced by the reduction
in its total numbers, and the comparatively high proportion of neutral space.

The arrival of the soybean aphid (Aphis gli'c'ines) in 2000 further illustrates the
importance of the temporal characteristics of these species. The soybean aphid arrived in
2000, and since its arrival, it has been a late season pest of soybean. The LTER has been
colonized later than most areas of the state, and it is possible that the high populations of
C. septempunctata aid in the delay of colonization. The spread of soybean aphid later in
the season is hindered further by the then high prevalence of H. axyridis. The overall
effect is fairly contiguous pest suppression mid season through harvest.

Without this temporal separation in niches the generalist behavior, lack of prey
specificity, and common habitat would bring these two species into direct competition
with each other, limiting one or both species’ ability to thrive. The principle of
competitive exclusion requires that two species vying for the same resource must
compete, and one of those species must eventually push the other out of the niche. This
push can be to less suitable prey to local extinction or into marginal habitat. This has
been shown to happen with H. axyridis, forcing it to rely on cannibalism (Snyder et a1
2000) or phytophagous resources (Slogget and Majerus 2000), but neither phenomenon
has been in evidence in relation to competition with C. septempunctata in the LTER
There is preliminary evidence that suggests that in times of limited prey availability in the
row crop treatments, the two species move into separate alternative habitats: H. axyridis
into the Populus Sp. and arboreal treatments, and C. septempunctata into old field and

early sucessional habitats. These observations, if shown to be significant by fiiture

27

analysis would further reduce the potential for competition between the species by spatial
separation.

These two species as fairly new arrivals to North America have been able to
establish populations and disperse widely; both have been successful invaders. The
arrival of H. arvridis years after C. septempunctata seemed to set the stage for a
competitive interaction and the eventual establishment of one species as the dominant
member of the community; however, the segregation temporally between the two has
allowed both to thrive. This phenomenon can be an effective pest management paradigm
allowing consistent control of aphid pests throughout the latter two thirds of the
Midwestern growing season. This control comes at a price; however, the rise in exotic
species may have come at the expense of less competitive native species such as Adalia
bipunctata L. which has not been observed in the LTER for several years (LTER
unpublished data). The decline of native species is a plausible result of the arrival and
establishment of exotics of the same guild and habitat preference. The costs to the
environment of establishing exotics, even those considered beneficial, should be carefully

weighed against their future benefits as biological control agents.

28

Chapter Four: The Microhabitat Preference and
Response of the Dominant Members of the Coccinellid
Complex to an Exotic Pest

Introduction:

The competitive exclusion principle states that two or more species can not
coexist indefinitely on the same resource. Classic experiments on competition, such as
those by GP. Gause in the 1920’s, have shown that one species can have a detrimental
effect on another (Gause 1934), but this impact in a natural setting is much harder to
quantify due to the spatial, temporal, and behavioral dynamics of the species niches. The
niche of a species is defined by Ricklefs (1997) as “the ecological role a species plays in
the community, and niche overlap is the sharing of niche space by two or more species;
similarity of resource requirements, and tolerance of ecological conditions.” To avoid the
competitive exclusion principle’s consequences, the niches of the species present in a
community must not overlap completely. The separation in niches can be temporal,
spatial, behavioral, or some other factor, but it must be sufficient to allow both species to
coexist indefinitely. These separations in the niches can be hard to identify, much less
quantify in a community, so the apparent absence of competition can be confounding.

The niches present in a community are also subject to outside influences such as
invasive species. The introduction and establishment of an invasive exotic either via
accidental or intentional introduction, represents a permanent alteration of the system.
There are numerous examples of harmful introductions of plants, animals, and pathogens
and their costly and damaging effects on native communities (Ruesink et a1 1995). The

entrance of an exotic into a niche previously filled by a native species, can drive the

29

native out due to the inherent advantages of being non-native. Exotic species are
generally free of their traditional predators and parasitoids, as well as their pathogens.
Invaders generally are habitat and prey generalists, reproduce quickly, and have a means
of rapid dispersal (Ricklefs 1997). The landscape level impacts of invasive species
resemble the recolonization of islands in Wilson’s classic experiments on island
biogeography (Wilson and Macarthur 1967). As a new species moves in, it experiences a
surge in numbers due to the advantages previously stated. As it establishes itself, if the
niche it moves into is filled, the older resident is forced out either into marginal
conditions or into local extinction. Later, after the invader has become firmly established,
its advantage starts to wane. This could be due to the arrival of predators and pathogens
or due to the adaptation of the ecosystem to its presence (Ricklefs 1997).

The continuing concern about the impacts of pesticide use and genetically
modified crops is keeping the seemingly benign practice of classical biological control at
the forefront of agricultural pest control. This practice seeks to control pest outbreaks by
the introduction or augmentation of predators. These predators are sometimes native, but
often the pest is an accidentally introduced exotic, so a predator from the same
geographic area as the pest is imported. This brings two exotics into the same system and
conflict with the endemic species in that system is often the result. Some of the earliest
and most successful biological control agents imported into the Unites States were
coccinellids (ladybeetles). They have been introduced numerous times throughout the
country to control pests in crops ranging from citrus to forests to row crop agriculture

(Gordon 1985).

30

Coccinellids are voracious predators of certain agricultural pests, especially
aphids. In agricultural settings the coccinellid predator community provides protection on
a variety of crops (Gordon 1985). This has led to their widespread use in biological
control applications. Species have been moved from place to place to control outbreaks of
endemic and exotic pests. The repeated introduction of exotic coccinellids (Harmonia
axyridis Pallas and Coccinella septempunctata (L.)) in biological control has led to their
establishment throughout the United States. C occinella septempunctata was introduced
after the 19505 and has become established in the northeastern United States (Schaefer et
al. 1987). The introduction of H. aryridis into the United States was either intentional or
accidentally via commerce (Day et al. 1994). Independent populations were first
documented in 1988 (Chapin and Brou 1991), since it has expanded its North American
range to become the most common ladybeetle in most of the United States. While these
species represent a good weapon against sofl bodied insect pests like aphids and mealy
worms there is also evidence that they are having deleterious impacts on endemic
populations of ladybeetles (Wheeler and Hoebeke 1995, Elliot et al. 1996, Colunga-
Garcia et al. 1998). These introductions and the subsequent establishment in Southern
Michigan have fundamentally altered the arthropod community ecology in the region.
The effects on native coccinellid community structure are largely unknown due to a lack
of knowledge about the community before the introduction of the exotics.

A diverse mix of urban, old field, riparian, wetland, and agricultural habitats
characterizes the Southern Michigan landscape. This habitat diversity is reflected in the
diversity of the arthropod communities. The high landscape diversity also provides

abundant resources for generalist predators. Harmonia arvria’is is adapted to arboreal

31

habitats, but as a generalist it is able to exploit resources in old field and agricultural
habitats as well. This flexibility allowed H. axyridis to become the dominant coccinellid
in Michigan within five years of its first detection (Colunga-Garcia and Gage 1997).
Coccinella septempunctata is prevalent primarily wheat, alfalfa, soybean, and old field
habitats (Honek 1985, Ostrom et al. 1997) and was the dominant coccinellid before the
arrival of H. axyridis. Even after the arrival of H. wryridis, C. septempunctata is still the
second most common coccinellid in Southern Michigan.

The first populations of the soybean aphid (Aphis glycmes Matsumura) in the
north central region were documented in the late summer of 2000. A broad range was
already established, as soybean aphids were documented in Illinois, Indiana, Iowa,
Kentucky, Michigan, Minnesota, Missouri, Ohio, West Virginia, and Wisconsin. The
heaviest infestations were in Michigan, southern Wisconsin, southeastern Minnesota,
northern Illinois, and northern Indiana (DiFonzo and Hines, 2001). China and Eastern
Asia are its traditional home range, and it is documented in the Philippines as well as
Australia.

The establishment of a new pest on a vital cash crop is a cause for concern. Like
all aphids, Aphis glycmes is a vascular feeder, piercing the outer leaf surface to extract
liquid form the vascular tissues of the soybean plant. Soybean is generally seen as a crop
tolerant to insect damage. However, Aphis glycines populations are capable of extremely
rapid reproduction and hundreds of individuals per leaf have been observed. These
population levels especially when coupled with dry weather can cause significant yield

loss (Ostile 2002). The soybean aphid is also capable of transmitting plant pathogens

32

such as soybean mosaic virus, although the extent of transmission in the United States is
unknown (Ostile 2002).

Understanding the impact of pest control introductions on endemic communities
is essential to making good long-term management decisions, however, without a
thorough understanding of the structure of the current community future impacts on
biodiversity, ecosystem function, and competitive displacement of endemic species
cannot be evaluated. For future decisions affecting coccinellid introductions a detailed
examination of the coccinellid communities ecology including, niche overlap,
competitive interactions, and preferred prey has to be done. Of the fourteen coccinellid
species present in the LTER the dominant four were Harmonia axyridis, Coccinella
septempunctata, Coleomegilla maculata (Degeer), and C ycloneda munda Say.

We hypothesized that the four dominant coccinellids would show segregation
based on habitat type, and that the community would respond to the presence of soybean

aphid by changing its dispersion patterns.

Materials and Methods

To begin drawing a detailed picture of coccinellid community structure
and function, a monitoring system was designed to assess landscape level microhabitat
preference among the four most prevalent coccinellids in Michigan, using the Kellogg
Biological Station (KBS) Long Tern Ecological Research Site (LTER). The objective of
this project is to describe the structure and function of the coccinellid community in an
agricultural landscape. KBS in Southwestern Michigan is representative of the habitat
types and cropping systems common to the region. In 1988 The National Science

Foundation established a LTER there with the purpose of assessing the impacts of

33

agriculture on biodiversity, community ecology, and ecosystem function. The site for
this study is located in ancillary plots adjacent to the LTER main site. The habitat is
highly fragmented, with row crops, perennial alfalfa populous, hardwood, and conifer
plots. The landscape diversity is high, and this leads to a high degree of arthropod
predator diversity. The LTER has been monitoring the coccinellid community for 13
years, assessing the impact of different land use regimes on the efficacy of endemic
biological control.

During the 2001 and 2002 growing seasons, coccinellid adults were sampled
using double-sided, yellow Pherocon AMTM un-baited cardboard sticky traps (22.5 x 14.0
cm). The traps were attached at plant canopy height, one meter, three meters, and five
meters above the soil surface to either “t” posts as described by Maredia et al. (l992a), or
affixed to fiberglass rods that were in turn attached to six foot steel fence posts. The
sampling stations were arrayed at the edge of the field then 45 m in six were placed at
intervals of 20m. The placement of the rows of monitoring stations was staggered to
leave a trap free space in the center of each plot to avoid “trapping out” the fields. The
second row of monitoring stations was placed 15m south of the first then a gap of 30m
was left before the next row (See photo with trap locations at the top of Figures 14-19).
This sampling regime provided a staggered grid where each row of sampling stations
serves as a sub sample of the plot it is located in, providing for two equal sub samples
nested within two replicates of the plots. Three hundred and fourteen traps in all were
placed in rows across the plots. The plots are all approximately 1.15 hectares each for a
total sampled area of 4.6 hectares. The plots were prepared as shown in Tables two and

three. The crops were rotated for 2002 keeping with standard agricultural practice.

34

 

Date: Action: Plot:

 

 

 

4-28-2001 Disced with eight inch disc Com one and two soybean one
and two

4-29-2001 Soil finished C orn one and two soybean one
and two

5-3-2001 Planted PioneerTM 92862 Soybean one and two

soybean at 150.000 seeds." acre in
30” rows (roundup ready)

 

5-4-2001 Planted PioneerTM 3730 corn at C om one and two
26,000 seeds/acre 30”rows while
applying 10 gallons/acre 28%

 

 

nitrogen

5-9-2001 Broadcast sprayed with C om one and two
Broadsroke+dual+atrizineTM

6-20-2001 Broadcast sprayed with Roundup Soybean one and two

UltraTM and ammonium sulfate

 

6-21-2001 Sidedressed with l IOIbs/acre of Com one
28% nitrogen

 

 

6-26-2001 Sidedressed with I 101bs/acre of C om two
28% nitrogen

 

 

 

 

Table 2 Preparation and planting regimens for the study plots in 2001.

The traps were replaced every second week. Coccinellid adults were identified, counted,
recorded, and removed every week. Over 10,000 individual specimens are represented in

this analysis.

 

Date: Action Taken: Treatment:

 

5-20 -2002 Soil finished all four plots

 

5-21-2002 Planted to pioneer 37m34 corn at 28,000 seeds per acre Com one and two

 

5-24-2002 Broadcast sprayed plots 27-A and 27-C with 6 ounces C om one and two
per acre callisto, 1 quart per acre atrazine and 1 quart
per acre dual

Planted pioneer 92b38 soybeans at 180,000 seeds per Soybean one and

 

 

 

 

 

acre two

6-27-2002 Sidedressed plots 27-A and 27-C with 1251bs of Com one and two
nitrogen or 42 gallons per acre of 28%
Broadcast sprayed with roundup ultra at 1 quart per Soybean one and
acre two

 

Table 3 Plot preparation and planting regimes in 2002
The habitat was blocked into three types, corn, soybean, and edge. Each block

contained 20 sampling stations. The edge habitat consisted of 2 m transects along the

corn soybean boundary. The edge areas were not prepped as the cropping systems were,

35

merely left alone. This produced a weedy strip between the crops and around the edges of
the entire plot. The plant community present in the edge areas consisted of a typical early
sucessional community with species like pigweed, lambsquarter, dandelion, thistle and

grasses. The diversity of the edge plots was as expected much higher than the crop

treatments according to Simpson indexes (D 21/2 Pi 2 ) of the three treatments

(Ricklefs 1997). Measured diversity was edge I 5.2, corn 2 1.4, soybean = 1.1.

The location of the sampling stations was determined with a global positioning
system then mapped in Arc GIS. The weekly trap catches were mapped by species and
also plotted according to habitat. The three and five meter level traps are not included in
this analysis to separate longer range movement from short term foraging flights.
Coccinellids tend to fly higher when dispersing farther and low or at plant canopy height
when actively searching for prey. The soybean aphid was sampled by counting the
number of aphids on five soybean leaves within 2 m of each sampling station. The aphid
count was then indexed in the following scale: 0:0, 1:1-10, 2:1 1-20, 3:21-30, 4:31-40,
4-41 and over.

The data was analyzed with SAS© to test whether the data met the assumptions of
ANOVA analysis then with a one way analysis of variance to show the effect of the
soybean aphid. Then a series of orthogonal contrasts to show the effect of increasing

aphid populations and habitat preferences was done. The statistical model for the

contrasts is y”. = ,u + alA“ + azAzl. + a3A3, + 014 A4,. + “5A5.- + e”. , where y” = predicted
predator density, ,u = the overall species mean, “1A“ + 512/42,- + a314,, + Q4144, + Q's/15,» = the

aphid scale values with their alpha values and, e”. = the residual. All a values = .05. This

model was used because it is ideal for examining repeated (in this case weekly)

36

measurements for treatment effect (Kuehl 2000). The model also partitions the treatment
sum of squares in ANOVA to minimize the effect due to the replicates. The results of the
orthogonal contrasts, where significant, were then used to establish predictive linear
models with the equation; y=mx+b. y: predator population mean, m = slope or the best
fit line, x = the aphid density and b = the x intercept. Least squares linear regression was
the best fit, with alpha 2 .05.

The preference of the species for a habitat was statistically tested using the
following linear orthogonal polynomial contrast model,
y”. = ,u +all-Il, + asz. +a3H3I. + e

where y”. = predicted predator density, p = the

a ’
species mean, aIH“ +a3H2, +a3H3, I the three habitat types and their alpha values and,
e”. = the residual. This model was used for the reasons stated above.

The aphid index was then interpolated using inverse distance weighting in ARC
GIS 8.1TM and displayed on the maps as contour lines. The coccinellid dispersions were
mapped with the same methodology and were displayed with color ramps to show
intensity. The legend is classified by standard deviation from the mean, with the darker
colors representing higher population densities. The resultant maps were then overlaid
onto a high resolution aerial photograph of the study site to give a perspective on the

surrounding habitat.

Results

Habitat Segregation

Each of the four dominant species was found in significant numbers in all

microhabitats. However, all four species showed a marked preference for a particular

37

 

 

 

 

 

 

 

 

 

 

Year Species Alpha Pr>F R 2

200 l C occirwllu .wptcmpmu'tutu .05 000 1 . 307
200' C'U/c’mm’gi/ 1U "law/aft! .05 .059 . 104
2001 H WW)" "0 “AWN/'3' .05 .0042 . 193
2001 C1‘('/0m’du mum/U .05 .0105 . l 64
2002 Coccinella yeprempuncrum _05 <,OO()1 , l 1 2
2002 C olemnegilla maculata .0 5 <,OOO] , 1 40
2002 Harmuniaaxyridis .05 <.0001 .134
2002 Oriana/a mum/a .0 5 . 1 336 .01 7

 

 

 

 

 

 

Table 4. 2001 ANOVA results and correlation coefficients for the response of the coccinellids to the
habitat type index for 2001 and 2002. If the Pr>F is less than .05 the species showed a significant

response to habitat. The correlation coefficient ( R 2) indicates the strength of this response.

habitat (Figures 14-17). The ANOVA (alpha 2 .05) analysis shows a significant
treatment effect for all three microhabitats as well as all four coccinellids (Table 4). The
more specific among the insects (Coccinella septempunctata) showed the highest R2
value, while the more generalist ones had lower corollary values.

The orthogonal polynomial contrasts also showed significant differences in
microhabitat for each species (Tables 5 and 6). Coccinella septempunctata showed
significance when contrasting com and soybean as well as soybean vs. edge, but not in
corn vs. edge. C. maculata showed significance when contrasting corn vs. soybean but
not com vs. edge or soybean vs. edge. H. axyridis showed significant differences when

contrasting corn vs. soybean and soybean vs. edge.

38

 

 

 

 

 

Contrast
C. C. H. C.
septempunctata maculata axyridis munda
Corn vs. X X X
Soybean
Corn vs. Edge X X
Soybean vs. X X X
Edge

 

 

 

 

 

 

Table 5. 2001 Linear orthogonal contrasts between treatments for each species in 2001. Contrasts
marked with an X are statistically significant using linear orthogonal contrasts. Significant contrasts
show a difference in habitat preferred.

 

 

 

 

 

Contrast
C. C. H. C.
septempunctata maculata axyridis munda
Corn vs. X X X X
Soybean
Corn vs. Edge X X
Soybean vs. X X X
Edge

 

 

 

 

 

 

Table 6. Linear orthogonal contrasts between treatments for each species in 2002. Contrasts marked
with an X are statistically significant using quadratic linear contrasts. Significant contrasts show a
difference in habitat preferred.

Each species average trap catch is graphed below by trap row (Figs. 14-17). As
the graphs below indicate C. septempunctata and H. axyridis are significantly more likely
to be found in soybean than corn. Corn is the overwhelming preference of C. maculata.
While C. munda is the only species studied that was predominant in the edge habitats;
this only occurred in 2001. In 2002 C. munda numbers were very low, and they showed
no significant difference between treatments. The averages by microhabitat correlate
well with theoretical models of habitat as shown in figure 20. While none of the three
species fits the true generalist pattern, Harmonia axyridis was displayed the most

flexibility in microhabitat selection and is shown with the generalist model.

39

Aphid Responses:

C accinel/a septempunctata preferred the soybean fields to the corn and was
sampled in large numbers in the early season (Appendix 2, Figure 23). It actively fed on
the soybean aphid, but its potential as a biocontrol agent is limited because in Michigan
soybean aphid is a late season pest, and C. septempunctata numbers are at their highest in
the middle of the growing season, tapering off quickly at approximately the same time as
the soybean aphid reached its population peak. C. septempunctata populations were high
in the early season and possibly had a direct impact by delaying soybean aphid
colonization early in the season. The research area was successfiilly colonized later than

most other areas in the state.

 

 

 

 

 

 

 

 

 

 

Year Species Alpha Pr>F R 2
200l C occ'inella septempunctata .05 .0042 .478
2001 C aleomegilla maculata .0 5 . 1403 .269
2001 Harmonia art'ridis .05 .0001 .6] 8
2001 Civc'loneda munda .05 .3816 .181
2002 Coccinella septampuncta(a .0 5 . 754 .030
2002 C aleomegilla maculata .05 .6393 .030
2002 Harmonia axyridis .0 5 .629 .032
2002 CI’C‘IUWdU mum/U .05 .966 .0023

 

 

 

 

 

 

Table 7. ANOVA results and correlation coefficients for the response of the coccinellids to the
soybean aphid index for 200] and 2002. If the Pr>F is less than .05 the species showed a significant

response to the aphid. The correlation coefficient ( R 2) indicates the strength of this response.

C yclonea'a munda preferred the edges and weedy strips between the crops before

the arrival of the soybean aphid. After the colonization of soybean aphid C. munda

40

slowly moved into the soybean fields. The dispersion into soybean seems to be a direct
response to the presence of soybean aphid, as its population in soybean rose linearly with
the rise of soybean aphid numbers. C. mzmda moved into the fields after soybean aphid
populations were high, helping to prevent an infestation (Appendix 2, Figure 22).

Harmonia arvridis is the most generalist of the coccinellids represented here. It is
considered primarily an arboreal species but is adept at exploiting a number of habitats
and prey. This is indicated in the maps by its continued sizeable presence in corn as well
as soybean. H. axyridis was by far the most common coccinellid in the study and
consequently was the main predator of the soybean aphid (Appendix 2, Figure 24).

Coleomegilla maculata was the least responsive to the presence of the soybean
aphid (Appendix 2, Figure 21). This species prefers corn, where it feeds on corn pollen,
mites and aphids in the tassels. It never moved in any appreciable numbers into soybean,
indicating a possible lack of behavioral plasticity (i.e. not recognizing the soybean aphid
as a food source).

The numerical response to the aphid was varied. The significant results of the
linear orthogonal contrasts (Table 8) show a marked response in only two species in 2001
and none in 2002. In 2001 only C. septempunctata and H. axyridis showed response to

the presence of the aphid.

41

 

 

 

 

 

 

 

 

 

 

 

 

Year Species Contrast Pr>F R2 Linear Model

2001 C. septempzmc'tata 0 vs] .005 .48 y: .79 x + .294
2001 C. septempunctata 0vs 2 .0097 .48 y: .79 x + .294
2001 C. septempunctata 0 vs. 3 .0004 .48 y: .79 x + .294
2001 C. septempunctata 0 vs. 4 .0097 .48 y= .79 x + .294
2001 C. septempunctata 0 vs. 5 .0170 .48 y: .79 x + .294
2001 H. axyridis 0 vs.1 .0003 .61 Y: 19x + 33.4
2001 H. agiridis 0vs 2 <.0001 .61 Y: 19x + 33.4
2001 H. axyridis 0 vs. 3 <.0001 .61 Y: 19x + 33.4
2001 H. axyridis 0 vs. 4 <.0001 .61 Y: 19x + 33.4
2001 H. wrvridis 0 vs. 5 <.0001 .61 Y: 19x + 33.4

 

 

 

 

 

 

Table 8 Results of the linear orthogonal contrast for 2001 aphid scale test. Only the significant

contrasts and models are shown.

None of the intermediate contrast levels (i.e. 2vs3 or 3v.s.4) for 2001 were significant but
zero contrasted with any level was significant for the two exotic species. The aphid scale

for 2002 was not significantly different than zero due to the low numbers of aphids so the

contrasts were all insignificant, and no predictive equations could be developed.

42

 

Figure 14: The averages per row for the native Coccinellids in 2001.
The study area is shown above the graphs with the corresponding row
averages as bars directly below the actual sampling row. Significant
differences in treatment are designated by letter. The Y axis is the
average weekly trap catch for the six trap stations in the center of the

rows. Images in this thesis are presented in color.

43

 

 

 

2001 Native Species Averages By Row?

 

  

 

Cycloneda munda by Trap Row

1
l
l A
‘ A
l
l
l
1

 

A
B B 8I B B B| B B
I I I I I I II

 

 

 

 

 

 

Coleomegilla maculata by Trap Row

A A ,
A

A A

A B B

BBBI ILi I

 

 

o 6

0 5

0.4

0.3

o 2
.C

3
N

U o
D.
N
h

I— 0.4

d) 0.35

g
h

0 0.25

> 0.2

< 0.15

0.1

0.05

0

Figure 14

44

Figure 15: The averages per row for the exotic Coccinellids in 2001.
The study area is shown above the graphs with the corresponding row
averages as bars directly below the actual sampling row. Significant
differences in treatment are designated by letter. The Y axis is the
average weekly trap catch for the six trap stations in the center of the
rows. Different letters indicate significant differences in response to

habitat. Images in this thesis are presented in color.

45

 

 

2001 Exotic Species Averages By Row I

    

a at)";

a

.:I|<.A°‘ 'Q,

 

 

-’ —' . ' . . ‘ I
‘ l r“ ’ ‘I, ‘ i u

"1 3'7“,-

1 A Harmonia axyridis by Trap Row
I A ‘ A
1 5 j A
1 B . .
l B B B B B B B
“1 11 11 ll 18
l
0 1 V 7 s . , V. fl . , a .. __ 7.1

Coccinella septempunctata by Trap Row
A A

 

 

 

 

 

Average Trap Catch

 

 

 

 

 

Figure 15

46

Figure 16: The averages per row for the native Coccinellids in 2002.
The study area is shown above the graphs with the corresponding row
averages as bars directly below the actual sampling row. Significant
differences in treatment are designated by letter. The Y axis is the
average weekly trap catch for the six trap stations in the center of the
rows. Different letters indicate significant differences in response to

habitat. Images in this thesis are presented in color.

47

 

2002 Native Species Averages By Row

 

 

 

Cycloneda munda by Trap Row

    

 

Average Trap Catch

 

 

 

 

 

 

Figure 16

48

Figure 17: The averages per row for the exotic Coccinellids in 2002.

The study area is shown above the graphs with the corresponding row
averages as bars directly below the actual sampling row. The Y axis is
the average weekly trap catch for the six trap stations in the center of
the rows. Different letters indicate significant differences in response

to habitat. Images in this thesis are presented in color.

49

 

1 2002 Exotic Species Averages By Row

    

 

 

 

Average Trap Catch

 

 

 

 

 

 

Figure 17

50

Figure 18. H. axyridis and C. septempunctata and aphid scale vs. date
for both years. The data points represent average trap catch at the
center six traps stations in each row for the Coccinellids and the
actual aphid scale sampled weekly. Error bars represent the standard

error. Images in this thesis are presented in color.

51

Average Trap Catch

Average Trap Catch

Figure 18
C. septempunctata and Aphid Scale vs. Date

  

 

 

+ C. septempunctata + aphid scale.1
3.5 ' '
3
25
2
1.5
1
0.5
O r—o» :‘SET‘ fin—WM:
.. _\\ \fxxxN’L’LfL’I’L’L'L'L'L
062363® :9" st"? «‘ng ‘69: 02-“? 6069 6\,\°6a.9&®0 ®NQNs§9s§90§ KG§Q@@96\QQG<\® 3°76 «”6 «Sce‘eoxflggxfi 64°
Date
H. axyridis and Aphid
Scale vs. Date
3 l
., '1
. ~— H. axyridis - aphid scale 1
2.5 1 g
2 .1/ \
l 1 1
l
l i 1 1
.11 1 1 1. 1 \
1 5 ,1, 1 11 l I", (i1
,1 1 1 1 \ 1 , 1 l
1‘ 1. 1 1 1 ( 11 1 \ 1.
1 /1 '. 1 l l ,1 /
/ 11 1‘ 1 / 1 1/ /
1i l 7 l .
1' 1 .R / i. ,1 «xx 1..
0.5 1 1 , . l/ I K/ \\ ' . ~ i
l /l‘ 4 f . \‘X X/ \\\ ,1/ \I‘ ’ I' V t
-. / .. . . \
1’ ‘ - . 8 ) a _.
0»‘,r~.--v-..or...rdr » .v-,v fihi" .. -o o 5 o o o o 07’:- .,,a
\ y\ ’x x '\ BK .'\ ’1' t} 5'} '1 ’1' fir ,"v
°®:.~*‘°..®‘°..<9‘°« .s 6* «41°69 q; “as" .0 “f a 9.5“ 6” 06°. a $960931... “5° e «91° saga???

Date

52

Figure 19. C. maculata and C. munda and aphid scale vs. date for both
years. The data points represent average trap catch at the center six
traps stations in each row for the Coccinellids and the actual aphid
scale sampled weekly. Error bars represent the standard error.

Images in this thesis are presented in color.

53

 

Figure 19
C. maculata and Aphid Scale vs.Date

 

 

 

 

0.8
. _ _ _ __ _
1 ' C. maculata + aphid scale 1
07 1: I
1
5 o.
.5 R
o 05 1 1.
a. r” \
N 04 ‘ T
i. ’ / \g I
1' T t 1. i
at 03 1 ‘ . 1.
a ‘1
x _
a; 02 I \\ I I I
2 1
01 z 3 I 1 \ ‘3 /“' ---a
.x 1 I \ 3‘ x I
1’ ‘ - \i q. ' ‘ 1
0 -—o —; o o.o.~o.4‘4———.———v———— —T——-—-—.——fi—— —.--..—c.-0~.-&T—o-To— . o---—--&.—L-r—w—-—1
‘0 \ ’\ ‘1 9'1, ’lv "Ir
0°e.~°e3°&«° «3° \3 N 3° 99°; 3° e“ 96139.9? 3‘ °We° «9° $996.0 s°6.s°6. 936$ .39. 00186390 (9° 49°: 3°
Date
C. munda and Aphid
0.. Scale vs. Date __ _ .__ _
1 + C. munda ~- aphld scale 1 1
0.4 “ E
035 1
c :
u 03 1
a
(B . .
o 025 1
Q 1
g 0.2 1
I- 1- . ‘
q, 0.15 \ . 1
a: 1 \ ’1 \ 1
e 0.1 1 1 1 1‘
0 ’- 11 k .
2 0 05 -\ 2"“ ‘~ /\ 1 1, \
,"1/11\\ '11”, 1‘ V. 1‘“ / :1, \ -- / 1
0 ._ .n-s #3:- _ -7_ __._ x -__.¢_ 3 _,_ T _... fll__.r._.._..__.._‘_.|
\e N K ’1' (16°
°° (19992599299 «‘36 @0606) 9Q ’9 966:9 "’0‘ 29 9&6 611$ 0°93?) \Nxoif9° 030.0 6° W‘éoe 9 W41 W49 3‘66: me

Date

54

Discussion

Habitat segregation

The results of the orthogonal contrasts provide a picture of habitat preference in
this system. C occinella septempunctata showed a significant difference in contrasts
between corn and soybean as well as soybean and edge. Coccinella septempunctata was

the most specific of the four species as represented by the highest correlation coefficient

( R2 =.307) clearly preferring soybean. The potential greatest potential competition threat
comes from H. axyridis but the two species’ niches are separated temporally not entirely
spatially (Chapter 3). C occine/la septempunctata is common early and in the middle of

the growing season but its population drops markedly in the late season each year.

In stark contrast to C. septempunctata, C. maculata shows a definite preference
for corn over both of the other treatments. It did utilize edge habitat but was
comparatively uncommon in soybean. The soybean aphid (Aphis glycmes) was abundant
and readily available in soybean in 2001, so the relative isolation of C. maculata in corn
indicates either a lack of behavioral plasticity on its part (failing to recognize the exotic
aphid as a food source), a failure to thrive when eating the soybean aphid, or that it
simply avoids competition with the larger more aggressive species in soybean. The
former of these possibilities seems the most likely, but they both should be examined
carefully to assess C. maculata ’3 potential as a biological control agent for Aphis

glycmes.

Cycloneda munda was the only species with a clear preference for the edge
microhabitat rather than either of the row crop alternatives although this preference was

only evident in 2001 because in 2002 its population was extremely small and no

55

significant differences in habitat could be detected. In 2001 it did move into soybean after
the soybean aphid became common, but even this dispersion was slow and originated
form the edge microhabitat. C. munda prefers the more diverse edge microhabitat for
reasons that are still unclear, perhaps it thrives in the increased vegetation diversity and
its attendant increased prey diversity, or simply it avoids interspecific competition as
none of the other species were as common, of preferred the edge habitat. The population
of C. munda was the smallest of the species in the study; this follows the general
assumption that it is primarily an arboreal species. The movement of C. munda into field
crops seems to occur only when an abundant food source is present, so the lack of
soybean aphid could explain its virtual absence in 2002. It has also possibly suffered
from competition with H. axyridis as the long term population trend for C. munda at the

KBS LTER since H. axyridis arrival has been negative (Colunga-Garcia and Gage 1998).

The most general microhabitat preference was that of H. axyridis. It was found in
all treatments concurrently with all other species. Even as the most generalist predator in
the study, it showed a marked preference for soybean over either of the two
microhabitats. H. axyridis was by far the most common of the coccinellids in this study
with a total trap catch of over 7000 individuals. Considered primarily an arboreal species
in its Japanese home range, it has in North America exploited a number of habitat types
and cropping systems. The appearance of H. mridis in the United States before the
soybean aphid created a novel scenario where an excellent biological control agent was

widely established before the pest arrived.

56

Models of Species Dispersion by Microhabitat

 

 

 

 

 

 

Habitat Neutral
Species Soybean
Generalist
‘o
a)
o.
o.
E
l—
% Corn Edge Sov Corn 1 Edge T Sov
:J
3 Corn Specialist Edge Specialist
'0.
5
Corn 1 Edge | Sov Com | Edge I Sov

N
u:

1. C. septempunctata I C. maculata Ic. munda I H. axyridis}

‘ soy corn 1 edge 1,

1 edge

A
—h UI N

Average Trap Catch +/_ Std Err

 

com

1
Year and Habitat

 

2001

Figure20: Theoretical models of species habitat preferences and actual results. Images in this thesis
are presented in color.

57

The theoretical dispersion models depicted in figure 20 are well represented by the four
species studied with the exception of the microhabitat generalist. This does not
necessarily mean that there are no generalists in the coccinellid guild. The presence of
Aphis glycmes as an abundant food source in soybean most likely drew the generalists
into soybean to take advantage of an easily exploitable resource. Harmonia axyridis is the
most generalist of the species studied but indicates a preference for soybean most likely
due to resource availability. The soybean specialist profile fits C. septempunctata due to
the temporal characteristics of the two species; the soybean aphid was a late season pest,
while C. septempunctata was at its highest densities in the early to mid season before the
arrival of the soybean aphid in the row crop.

The tendency for the coccinellids to isolate themselves from each other spatially
minimizes the potential for direct competition between the adults of the community,
allowing for seemingly high niche overlap. The overlap in spatial distribution occurred
mainly in areas of abundant Aphis glycmes colonies. With an ample food supply, the
adults of the three species found in soybean did not appear to compete directly to the

point of interspecific population regulation.

58

Aphid response

The soybean aphid was detected the week of July 23 in 2001 and during the week
of July 28, 2002. These dates correspond to roughly 750 degree days (50° F. The aphid
numbers in 2001 were much higher than in 2002, most probably due to the wetter, cooler
summer in 2002. Aphis glycmes is susceptible to fungal pathogens (Ostile, 2002), and the
higher moisture in 2002 could have reduced aphid populations in 2002.

The most effective predator of the soybean aphid in the soybean was H. axyridis
(Appendix 2, Figure 24) The temporal dynamics of both species dictate that they reach
their population peaks at roughly the same time, both in the late season. The soybean
aphid population in 2001 was rising until large numbers of H. axyridis arrived in the
fields. The slow colonization of the soybean aphid indicates another possible control
point in the system. The soybean aphid spends three generations on its overwintering
host, buckthom, before moving into the soybean fields. This initial population may have
been regulated by the presence of C. septempunctata or one of the more arboreal
community members such as C. munda or other members of the genus Coccinella
attacking the soybean aphid on its winter host would reduce the number of winged adults
that eventually moved into soybean and thereby be an important control point. More
monitoring of the predators in buckthom is needed to quantify the predator response on

that host plant.

59

Chapter Five: Summary

Objectives

The four species studied showed a remarkable ability to segregate their niches to
avoid competition as adults. The means of niche separation varied including temporal,
habitat, and seemingly behavioral mechanisms. To summarize, objectives stated in
chapter one are restated followed by a summary of the research findings:

Objective One: To gain a better understanding of the mechanism that allows the
exotic species to coexist in the same habitat.

l. The exotic coccinellids were found in preliminary statistical analysis to
not be competing. The spatial and temporal analysis of their habitat use
shows that the mechanism of their niche separation is a temporal shift
between the adults of the two species, with C. septempunctata dominating
the early season and H. axyridis the late.

Objective Two: To gain a better understanding of the temporal, spatial, and habitat
characteristics that separate the niches of the four dominant members of the
community.

I. The four species were found to have varying mechanisms of niche
separation. C. maculata isolates itself through use of corn as a primary
habitat; it is the only coccinellid in the group to do so. Its ability to exploit
a habitat the others can not is augmented by its ability to use pollen,
including com pollen, as an alternate food resource. It also showed no

response to the soybean aphid.

60

2. Cvcloneda mum/a also was present throughout the growing season and

avoided competition by using an alternate habitat; thriving in the edges
between the row crops and venturing into them only when food was
abundant. When food was limited in the row crops, it remained in the
edges or possibly stayed in the trees reflecting its arboreal nature.
Harmonia axyridis was found in all habitats but only in great numbers late
in the growing season. This allows it to dominate the landscape in late
summer. As a habitat generalist, it was able to exploit varying resources in
seemingly opportunistically.

C occinella septempunctata was most prevalent in the early season, mating
and laying eggs before the other three species. This early season
dominance coupled with its marked preference for soybean over edge or

corn kept it in areas not utilized by the other species.

Objective Three: To gain a better understanding of the effects of an exotic prey

species in the form of the newly arrived soybean aphid on the four dominant members

of the community.

1.

Coleomegilla maculata showed no significant response to the soybean
aphid. It maintained its strong preference for corn in 2001 when the aphid
was abundant as well as in 2002 when the aphid was scarce.

C yc'lanea'a munda displayed an opportunistic ability to exploit the aphid as
a resource. In 2001 when aphid populations were high, it moved into the
soybean treatments In 2002 when the aphid populations dropped off, it

remained in the edge treatments or possibly in the surrounding woods.

6]

3. Harmonia ari'ridis was the most effective predator of the soybean aphid.
It showed both a numerical and spatial response to the presence of the
aphid at high of low densities. Its temporal characteristics match the
aphid’s well, with both populations peaking in the latter part of the
growing season.

4. Coccinella septempunctata showed a strong preference for soybean, but its
temporal characteristics made it an inefficient predator of the soybean
aphid. The peak of C. septempunctata ’3 population is too early in the
season for it to be a large impact on the population of the soybean aphid.
The high early and midseason populations could have played a significant
role in delaying the colonization of the aphid in 2001, while in 2002 the
very high numbers of c. septempunctata may have contributed to the
marked decline in aphid colonies in 2002.

5. The soybean aphid did not cause a major shift in the spatial or temporal
patterns of the dominant members of the coccinellid community. C.
munda was the only species to show a marked spatial response to the

soybean aphid.

Competition:

The debate about the role of competition in communities is referenced in chapter
two, and a brief summary of the role of competition in this community is relevant here.
The role of interspecific competition in this community currently seems minimal. This
lack of competition in the community could be due to the results of past interactions

which served to segregate the niches or a segregation by traits evolved independently of

62

the other species. The recent arrival of two of the coccinellids studied would seem to
favor the latter explanation due to the short amount of time the community has been
given to adjust to the exotics. The addition of anovel food source in the form of the
soybean aphid, a novel and rare occurrence, failed to elicit a response in any of the four
species high enough to cause direct intraspecific competition. This research also sheds no
light on the possible competitive interactions of the larval stages of these species. It is
possible that the niche segregation displayed by the adults is a direct result of competition

between the larvae.

Future directions:

There are several future research projects and data analyses that will further our
knowledge of this community and its dynamics. The first is to analyze all 13 years of the
LTER dataset to determine and quantify effects of different agricultural management
practices on the coccinellid community. As previously stated the treatments range from
certified organic to intensive conventional management. The analysis would use many of
the tools (both spatial and statistical) already used herein. The effects of inputs on
beneficial insects could play an important role in management decisions.

Second, I would like to test the effectiveness of C. septempunctata and C. munda
control of the soybean aphid on its winter and spring host. Buckthom is currently not
monitored at the LTER for either the soybean aphid or coccinellids. The quantification of
predation on the alternate host could provide an opportunity to interdict the pest before it
moves into the crop. This research would require the identification of buckthom
surrounding the study site, then sampling with yellow sticky traps as in Maredia (1992),

and developing an efficient means of sampling the aphid on this host.

63

Third, personal observations indicate that the application of Roundup® in
genetically modified soybean in mid June had a temporary negative impact on the
coccinellid community. The destruction of the weeds under the soybean killed up to one
third of the vegetation in the field at that time. The coccinellid numbers declined
precipitously for a short time. The potential side effects of this predator decrease include
a pest outbreak. The genetically modified soybean would be monitored for three weeks
pre and post herbicide application to quantify the population decline. For a control, a
soybean field would be planted with the same variety, and a tractor and sprayer pass
without the application of herbicide would be done at the same time the treatment fields
were sprayed. The weed community and the potential coccinellid prey would be
monitored for the entire period, as well as potential pests of the soybean.

The last proposed project is to quantify the temporal, spatial, and behavioral
dynamics of the larvae of the dominant coccinellids in the field. The adult populations
could be strongly influenced by interspecific egg predation, larval predation, and

cannibalism of both specific and con-specific larvae (Snyder, 2000).

64

References Cited

Baker, J ., French, K., Whelan, R. J. 2002. The edge effect and ecotonal species: bird
communities across a natural edge in Southern Australia. Ecology 83:3048-3059

Berg, H. van den, Ankasah, D., Muhammad, A., Rusli, R., Widayanto, H.A., Wirasto,
H.B., and Yully, I. (1997) Evaluating the role of predation in population fluctuations

of the soybean aphid Aphis glycines in farmers' fields in Indonesia. Journal of
Applied Ecology. 34(4): 971-984.

Blackman, R.L., Eastop, V.F. and Kibby, GO. (1997) "Taxakey: Aphids on the World's
Cr0ps." CD-ROM. CAB International & The Natural History Museum. ISBN 0
85199 172 6

Bogdanov, L. V. and Gagal’chii, N. G. 1986. Intraspecific variation in the Asian ladybug
Harmonia axyridis Pall. Near Vladivostok. Soviet Journal of Ecology 17: 108:1 13

Burbank, D. H., Pregitzer, K. S., and Gross, K. L., 1992. Vegetation of the WK. Kellogg
Biological Station. Michigan State University/ Agricultural Experimental Station,
Michigan

Chang,Y.D., Lee, J .Y. and Youn, Y.N. (1994) Primary parasitoids and hyperparasitoids
of the soybeanabean aphid, Aphis glycines (Homoptera: Aphididae). Korean Journal
of Applied Entomology 33(2): 51-55

Chapin, J. B., Brou, V. A. 1991, Harmonia avg/ridis (Pallas), te thire species of the genus
to be found in The United States (Coleoptera, Coccinellidae). Proc. Entomol. Soc.
Wash. 93: 630-635

Colunga-Garcia, M., Gage, S. H. 1998. Arrival establishment and habitat use of the
multicolored asian ladybeetle (Coleoptera: Coccinellidae) in a Michigan landscape.
Environmental Entomology 27: 1574- 1580

Colunga-Garcia, M., Gage, S. H., Landis, D A. 1997. Response of an assemblage of
Coccinellidae (Coleoptera) to a diverse agricultural landscape. Environmental
Entomology 26: 797-804

Colunga-Garcia, M. 1996. Interactions between landscape structure and ladybird beetles
(Coleoptera: Coccinellidae) in field crop agroecosystems. Thesis (Ph.D.)--Michigan
State University. Dept. of Entomology

Connell, J. H. 1980 Diversity and the coevolution of competitors, or the ghost of
competition past. Okios 35: 131-138

65

C ottrell, T. D., Yeargan, K. V. 1998. Intraguild predation between an introduced
ladybeetle, Harmonia axyridis (C oleoptera: Coccinellidae), and a native ladybeetle,
C oleomegil/a maculata (Coleoptera: Coccinellidae). Journal of the Kansas
Entomological Society 7 l : 159-163

Darwin,-Charles-Robert,; I926. The origin of species / Charles Darwin; abridged and
introduced by Philip Appleman. 1st ed. New York : Norton, [1975]

Day, W. H., Prokrym, D. R. Ellis, D. R., and Chianese R. J. 1994. The known distribution
of the predator Propyela quatuordecimpunctata (Coleoptera, Coccinellidae) in the
United States, and thoughts on the origin of this species and five other exotic
ladybeetles in eastern North America. Entomol. News 105:244-256

DiFonzo, C. and Hines, R 2001. Soybean Aphid in Michigan. Michigan State University
Extension Bulletin E-2748. 4 pp.

Drestadt, S. H., Hagen, K. S., Bezark, L. G. 1995. Harmonia axyridis (Pallas)
(Coleoptera: Coccinellidae), first western United States record for this Asiatic
ladybeetle. Pan-Pacific Entomology 71: 135-136

Ferber, D. 1999 Risks and benefits: GM crops in the crosshairs. Science 286: 1662-1666
Gause, G. F. (1934), The Struggle for Existence. Williams and Williams, Baltimore

Gordon R. D. 1985. The Coccinellidae (Coleoptera) of America North of Mexico.
Journal of the New York Entomological Society. 93:]

Hesler, L. S., Kickhefer, R. W., Becc, D. A., 2001. First record of Harmonia axyridis
(Coleoptera: Coccinellidae) in South Dakota and notes on its activity there and in
Minnesota. Entomology News l 12: 264-270

Hirano, H., Honda, KI. and Miyai, S. (1996) Effects of temperature on development,
longevity and reproduction of the soybean aphid, Aphis glycines (Homoptera:
Aphididae). Applied Entomology and Zoology. 31(1): 178-180.

Honek, A. 1985. Habitat preferences of apidophagous coccinellids (Coleoptera).
Entomophaga 30: 253-264

Howarth, F .G. 1991. Environmental impacts of classical biological control. Annual
Review of Entomology. 36: 485-509

IPM of Alaska. 2003. Asian Ladybeetle, Harmonia axyridis. Technical Bulletin:
Biological control series.

Katosyannos, P., and Aliniazee, M. T. 1998. First record of Strongygaster triangulifera
(Loew) (Diptera: Tachinidae) as a parasitoid of Harmonia axyridis (Pallas)

66

(Coleoptera: Coccinellidae) in western North America. Canadian Journal of
Entomology 130: 905-906

Krafsur, E. S., Kring, T. J., Miller, J. C., Naribolu, P., Obryvki, J. J., Ruberson, J. R.,
Schaefer, P. W. 1997. Gene flow in the exitic colonizing ladybeetle Harmonia
axyridis in Norht America. Biological Control 8: 207-214

Kuehl, R. 0., 2000. Design of experiments: Statistical principles of research design and
analysis. Brooks/Cole. Pacific Grove, California

Lamana, M. L., and Miller, J. C. 1998. Field observations on Harmonia axyridis Pallas
(Coleoptera: Coccinellidae) in Oregon. Biological Control 6: 232-237

Lyon, W. F. 2002. Ladybeetle. Ohio State University extension fact sheet HYG-2002-98

Magurran, A. E., Ecological diversity and its measurement. Princeton University Press.
Princeton NJ.

Maredia, K. M., Gage, S. H., Landis, D. A., and Wirth, T. M. l992a. Visual response of
Coccinella septempunctata (L.), Hippodamia parenthesis (Say), (Coleoptera,
Coccinellidae), and Chrysoperla carnea (Stephens) (Neuroptera, Chrysopidae) to
colors. Biological Control, 2: 253-256
1992b Ecological observation on predatory coccinellidae (Coleoptera) in
southwestern Michigan. Great Lakes Entomologist. 25: 265-270

Maredia, K. M., Gage, S. H., Scriber, J .M. 1992. Habitat use patterns by the seven-
spotted lady beetle (Coleoptera: Coccinellidae) in a diverse agricultural landscape.
Biological Control 2: 159-165

McClure, M. S. 1987. Potential of the Asian predator, Harmonia axyridis Pallas
(Coleoptera: Coccinellidae), to control Maesucoccus resosae Beau and Goodwin
(Homoptera: Margarodidae) in the United States. Environmental Entomology] 16:
224-230

McClure, M. S. 1986. Importing ladybird beetles to control red pine scale. Frontiers of
plant science, Connecticut agricultural Experimental Station 39: 5-7

Miller, M., and Applet, G. 1993. Biological control: a little knowledge is a dangerous
thing. Rutgers Law Review. 45: 285-334

Nalepa, C. A., Kidd, K.A., and Hopkins, D. I. 2000. The multicolored asian ladybeetle
(Coleoptera: coccinellidae): orientation to aggregation sites. Journal of the

Entomological Society of Georgia 35: 151-157.

Okamoto, H. Sato, M. 1973. Laboratory studies on feed ecology of adaphagous
coccinellids. Konchu 41: 342-348

67

Obata, S. 1986. Mechanisms of pry finding in the adaphagous ladybird beetle, Harmonia
axyridis (Coleoptera: Coccinellidae). Entomophaga 31: 303-31 1

Osawa, N. and Nishida, T. 1992. Seasonal variation in elytral colour polymorphism in

Harmonia axyridis (the ladybird beetle): the role of non-random mating. Heredity 69
297-307

Ostile, K. 2002. Managing soybean aphid. University of Minnesota Extension

Ostrom, H. P., Colunga-Garcia, M., and Gage S. H. 1997. Establishing pathways of
energy flow for insect predators using stable isotope ratios: fieldand laboratory
evidence. Oecologia 109: 108-1 13

Pecan-South I993, Ladybugs take commercial flight to Texas. Pecan-South 26: 2-3

Preziosi, R.F., Snyder, W. E., Grill, C.P., Moore, A. J. 1999. The fitness of manipulating
phenotypes: implications for studies in multivariate selection. Evolution 53 (4) 1312-
1318.

Ramakrishan, P. S. 2000, An intergrated approach to land use management for
conserving agroecosystem biodiversity in the context of global change. International
Journal of A gricultural Resources, Governance and Ecology l

Ricklefs, R. E. 1997. The Economy of Nature Fourth ed. New York: W. H. Freeman and
Company

Ruesink J. L., Parker 1. M., Groom M. J, Kareiva P. M. 1995. Reducing the risks of
nonindigenous species introductions. Bioscience 45: 465-477

Schaefer, P. W., Dysart R. J ., and Specht, H. B., 1987. North American Distribution of
Coccinella septempunctata (Coleoptera, Coccinellidae) and its mass appearance in
coastal Deleware. Environmental Entomology. 16: 368-373

Schoener,-T.W. 1997. Field experiments on interspecific competition [Plant and animal
associations]. American Naturalist. 122 (2) p. 240-285.

Scriber J. M. 200]. Bi or not Bt: Is that the question? Proceedings of the National
Academy of Science 98 (22) 12328-12330

Sisk, T. D., and Margules, C. R. 1993. Habitatedges and restoration: methods for
quantifying edge effects and predicting the results of restoration efforts. Nature
conservation3: reconstruction of fragmented ecosystems. Saunders, A., and Hobbs, R.
J. editors. Surrey Beatty andSons, Chipping Norton, New South Wales, Austrailia.

68

Sloggett, J. J ., Majerus, M. E. N. 2000. Habitat preferences and diet in the predatory
Coccinellidae (Coleoptera): an evolutionary perspective. Biological Journal of the
Linnean Society 70: 63-88

Snyder, W.E., Joseph, S. B., Preziosi, RE, Moore, A.J. 2000. Nutritional benefits of
cannabilism for the ladybird beetle Harmonia arvridis (C oleoptera: Coccinellidae)
when prey quality is poor. Environmental Entomology 29:1 173-1 179.

Southwood, T. R. E., 1978. Ecological methods. Chapman and Hall, London

Takahashi, S., lnaizumi, M. and Kawakami, K. (1993) Life cycle of the soybeanabean

aphid Aphis glycines Matsumura, in Japan. Japanese Journal of Applied Entomology
and Zoology.37(4): 207-212.

Tedders W. L., Schaefer P. W. 1994. Release ans establishment of Harmonia axyridis
(Coleoptera, Coccinellidae) I the southeastern United States. Entomol. News105: 228-
243

Thomas, M. B., and Willis, A. J. 1998. Biocontrol- risky but necessary? Trends in
Ecology an Evolution. 13: 325-329

Toumiaire, R., Ferran, A., giuge, L., Piotte, C., Gambier, J. 2000. A natural flightless

mutation in the ladybird, Harmonia axyridis]. Entomology Experimental and Applied
96 33-38

Ueno, H., Sato, Y., Tschida, K. 1998. Colour-associated mating success in a polymorphic
ladybird beetle, Harmonia axyridis. Functional Ecology 12: 757-761

Wang, C.L., Siang, N.I., Chang, GS. and Chu, HF. (1962) Studies on the soybean aphid
Aphis glycines Matsumura. A cta entomologica Sinica 1 1: 31-44.

Wang, S.Y., Bao, X.Z., Sun Y.J., Chen, R.L. and Zhai, B.P. ( 1996) Study on the effects
of the population dynamics of soybeanabean aphid (Aphis glycines) on both growth
and yield of soybeanabean. Soybean Science. 15(3): 243-247.

Wang, Y.Z. and Ba, F. (1998) Study on the optimum control of soybean aphid. Acta
Phytophylactica Sinica. 25(2): 151-155.

Wheeler, A. G., and Hobecke, E. R., 1995. Coccinella novemnotata in northeastern North
America: historical occurrence and current status (Coleoptera, Coccinellidae). Proc.

Entomol. Soc. Wash., 97: 701-716

Wilson, E. O. MacArthur, R. H., 1967. The theory of island biogeography.
Princeton, NJ. : Princeton University Press

69

Yasuda, H. and lshikawa, H. 1999. Effects of prey density and spatial distribution on
prey consumption of the adult predatory ladybeetle beetle. Journal of Applied
Entomology 123: 585-589

Yasuda, H., Kikuchi, T., Kindlmann, P., Sato, S. 2001. Relationships between attack
rates, cannibalism, and Intraguild predation in larva of two predatory ladybirds.
Journal of Insect Behaviour 14: 373-384

 

70

Appendix 1
Record of Deposition of Voucher Specimens*
The specimens listed on the following sheet(s) have been deposited in the named museum(s) as
samples of those species or other taxa, which were used in this research. Voucher recognition
labels bearing the Voucher No. have been attached or included in fluid-preserved specimens.

Voucher No.: 2003—03

Title of thesis or dissertation (or other research projects):

Quantifying the Roles of Competition and Niche Separation
in Native and Exotic Coccinellids, and the Changes in the
Community in Response to an Exotic Prey Species

Museum(s) where deposited and abbreviations for table on following sheets:

Entomology Museum, Michigan State University (MSU)

Other Museums:

lnvestigator’s Name(s) (typed)
Charles McKeown
Stua_rt Gaqe

Manuel Colun Garcia
Date S’Z/r 212293
*Reference: Yoshimoto, C. M. 1978. Voucher Specimens for Entomology in North America.
Bull. Entomol. Soc. Amer. 24: 141-42.

 

Deposit as follows:
Original: Include as Appendix 1 in ribbon copy of thesis or dissertation.

Copies: Include as Appendix 1 in copies of thesis or dissertation.
Museum(s) files.
Research project files.

This form is available from and the Voucher No. is assigned by the Curator, Michigan State
University Entomology Museum.

71

Appendix 1.1

Voucher Specimen Data

1 Pages

1of

 

new.“ is?“

3.28.5

  

m c Eo=2 9: :_ gmoaou

 

381%} 80

 

m_o.mo-md::_oo 628.2

 

ommo thw

 

.2 mcoEBoom BE. 9on on. oozooom

 

 

Page

 

5.63.22 8:ch

 

 

 

 

 

 

 

 

 

 

8.88 .oz 882$ 8...... @882 39888...
3.838: g Eoozm Ecosfium own.
8 so... R. .55 .mmv. 888.8. ._s_ 8.8... 8.88 2.3.x
F no... 8 ES «.9. 888.8. ._s. 5.88 8.5.: 8.88.80
F 8... 8 mm: 98. 888.8. .5. F33 8.5... 8.88.96
F no... 8 am: we. 888.8. ...2 F8: 8.5... 8.88.86
F no... 8 mm: m8. 888.8. ._s. 5.24 88.... 8.88.96
a 8.0... 8 mm: we. 888.8. .5. 8-8; 838... 8.8880
F .8... 8 mm: we. 888.8. .5. 3-8-.. 838.... 8.88828
8 no... 8 mm: .08. 888.8. ._s. 5.88 88:88.88 8.8.88
F no... 8. mm: m8. 888.8. ._.2 8-8: 8888.88 8.8.88
8 ea... 8 5.... 98. 888.8. ._s_ Fo-oF-F 8888.88 88.880
F no... 8 mm: mmv. 888.8. .__2 6-8; 8&8 8.8.8:
8 no... 8 mm: m8. 888.8. ..s. 518A 858 8828:
F 8... 8 am: we. 888.8. ._s. 888 888 8.8.8:
n no... 8 mm». we. 888.8. .5. 3-88 8.88 8.8.8:
d
8 0+ S
m m wen. r Psi m w .m. m s 8.682. 98 :92“... 850 .o 8.0on
w M W m u u m. m. w % com: .o 3.00:8 £8.50on .8 Emu .33
M w m 0 M .Ao P N m E

 

 

co .onEaz

 

 

72

Appendix 2 Soybean Aphid Response Maps

 

 

Figure 21 C. maculata

73

 

Figure 21 Cont. C. maculata

74

 

Figure 21 Cont. C. maculata

75

 

igure 21 Cont. C. maculata

F

76

. fl
1'
A}

 

Figure 21 Cont. C. maculata dispersion maps with the aphid scale
superimposed as contours for 2001. As the contour lines become closer
the aphid intensity is higher. The predator is shown by standard
deviation classification to emphasize the variation in the population
spatially. Images in this thesis are presented in color.

77

 

Igure 22 C. munda

F

78

 

munda

C.

Figure 22 Cont

79

 

Figure 22 Cont. C. munda

80

 

 

munda

C.

22 Cont

igure

F

81

 

Figure 22 Cont. C. munda dispersion maps with the aphid scale
superimposed as contours for 2001. As the contour lines become closer
the aphid intensity is higher. The predator is shown by standard
deviation classification to emphasize the variation in the population
spatially. Images in this thesis are presented in color.

82

 

 

 

Figure 23 C. septempunctata

83

,v

' B 9 -K- -. rar- ~ .2: -F.:.. ”A”; V
v '1' 1
1 ; ’V m m ‘; ‘
- ~ 1 It.
_ i I I
._ ’ g . ‘ _ {A A - I’
I . ‘-W ‘V ‘
‘ xx"
' a)" " '- i} )
Jul 1
\ ' ‘ ‘
J ‘

 

 

 

 

Figure 23 Cont. C. septempunctata

84

 

 

 

 

 

 

Figure 23 Cont. C. septempunctata

85

 

 

 

 

 

 

Figure 23 Cont. C. septempunctata

86

 

Figure 23 Cont. C. septempunctata dispersion maps with the aphid scale
superimposed as contours for 2001. As the contour lines become closer
the aphid intensity is higher. The predator is shown by standard
deviation classification to emphasize the variation in the population
spatially. Images in this thesis are presented in color.

87

 

 

 

 

 

Figure 24 H. axyridis

88

 

 

 

Figure 24 Cont. H. axyridis

89

 

 

 

 

 

 

Figure 24 Cont. H. axyridis

90

A.

 

 

 

 

 

Figure 24 Cont.H. axyridis

91

 

 

 

Figure 24 Cont. H. axyridis dispersion maps with the aphid scale
superimposed as contours for 2001. As the contour lines
become closer the aphid intensity is higher. The predator is
shown by standard deviation classification to emphasize the
variation in the population spatially. Images in this thesis are
presented in color.

92

Appendix 3: Height Graphs

 

Images in this thesis are presented in color.

2001 5m vs time

   

 

 

0.7 -
+ C. septempunctata + C. maculata
06 ~ C. munda —*- H. axyridis
05
3
t! 0.4
3
g
g 0.3
U)
02
01
0 I——- -——I—-—-—-—-I—+~-+ .
o" o" o " " o" o" o" o" o" o" o o o o o o
as 6.3} 60"} ,8 go. ,8 02‘ «<56 Ge (9} 619‘ 999 .p‘ Re $3 0“ ‘68 a
Date
2002 5m vs time
0.25 - - -..------..- , — —- ., ‘--— -— l
+ C. septempunctata + C. maculata
C. munda + H. axyridisg g g 1 ‘
0.2
g, 0.5 ?
E 5
‘ l
I; —\
i C7
0) 0.1 i
l
__———....x\
0.05 \ / 1
\fl_____ / I
A ‘/\‘|
0 L. . . . . . . . .F . 8
6/10/02 6/17/02 6/24/02 7/1/02 7/3/02 7/15/02 7/22/02 7/29/02 3/5/02 8112/02

Date

93

 

Speclee Average

Species Average

20013m vs time

o.a—-—«———-A— —--~ .. --

1 + C. septempunctata + C. maculata

l C. munda

+ H. axyridis l

 

 

 

 

\ K N N N N N N K K N N '\ K N K N
89 88° .48 «8° «8° «8° «8° «8° 28° 88° 49° 399 98° 98° 8‘9 48° 8° 88°
Date
2002 amveflme
0.6 . .. .7.-. - 7F -. F¥
+ C. septempunctata + C. maculata
05 C. munda + H. axyridis

 

4102 "
6/02 ‘
8/3/02 ‘

811/02

\\\\,,_._
\\\\\\

7
7

\\\\\

N N w
o o 9
B B o
N Q‘ Q
3 o to

94

 

8/5/02 ‘

 

6I1 3(02

 

 

 

2001 Height Averages by Species

 

 

 

 

 

 

 

Trap Height (m)

1.8 -_ — - - - -—-— -- --—--
1: l ~
8 1:2 ‘eAxgorc7 ]
S 1 ‘lAngfcmundi
E, 0-8 EIAngfCMAC
2 0'6 liAngfl-laxy
g 0.4 i --- <
< 0.2
o
o 1 3 5
Height (m)
2002 Height Averages by Species
l l
c l
l 2 l l
8 i ilAngfC? ll
E I lAngfcmundi-
\ 8 i iElAngfCMAC‘
E I ‘uAngfl-laxy J"

95