SPATIAL AND TEMPORAL ANALYSES OF BARK BEETLE POPULATION
DYNAMICS IN SOUTHERN BRITISH COLUMBIA: STAND-LEVEL STUDIES OF
THE BOLE-INFESTING ASSEMBLAGE DURING ERUPTIVE TRANSITIONS OF
MOUNTAIN PINE BEETLE, DENDROCTONUS PONDEROSAE HOPKINS

by

Jordan Matthew Koopmans
B.Sc, University of Northern British Columbia, 2008

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
NATURAL RESOURCES AND ENVIRONMENTAL STUDIES
(BIOLOGY)

THE UNIVERSITY OF NORTHERN BRITISH COLUMBIA
January 2011

©Jordan Matthew Koopmans, 2011

1*1

Library and Archives
Canada

Bibliotheque et
Archives Canada

Published Heritage
Branch

Direction du
Patrimoine de I'edition

395 Wellington Street
OttawaONK1A0N4
Canada

395, rue Wellington
Ottawa ON K1A 0N4
Canada
Your file Votre reference
ISBN: 978-0-494-75126-8
Our file Notre r6f6rence
ISBN: 978-0-494-75126-8

NOTICE:

AVIS:

The author has granted a nonexclusive license allowing Library and
Archives Canada to reproduce,
publish, archive, preserve, conserve,
communicate to the public by
telecommunication or on the Internet,
loan, distribute and sell theses
worldwide, for commercial or noncommercial purposes, in microform,
paper, electronic and/or any other
formats.

L'auteur a accorde une licence non exclusive
permettant a la Bibliotheque et Archives
Canada de reproduce, publier, archiver,
sauvegarder, conserver, transmettre au public
par telecommunication ou par I'lnternet, preter,
distribuer et vendre des theses partout dans le
monde, a des fins commerciales ou autres, sur
support microforme, papier, electronique et/ou
autres formats.

The author retains copyright
ownership and moral rights in this
thesis. Neither the thesis nor
substantial extracts from it may be
printed or otherwise reproduced
without the author's permission.

L'auteur conserve la propriete du droit d'auteur
et des droits moraux qui protege cette these. Ni
la these ni des extraits substantiels de celle-ci
ne doivent etre imprimes ou autrement
reproduits sans son autorisation.

In compliance with the Canadian
Privacy Act some supporting forms
may have been removed from this
thesis.

Conformement a la loi canadienne sur la
protection de la vie privee, quelques
formulaires secondaires ont ete enleves de
cette these.

While these forms may be included
in the document page count, their
removal does not represent any loss
of content from the thesis.

Bien que ces formulaires aient inclus dans
la pagination, il n'y aura aucun contenu
manquant.

1+1

Canada

ABSTRACT
Factors that trigger population transitions of mountain pine beetle from endemic to
incipient-epidemic levels are poorly understood. The population dynamics of this insect may
be influenced by associations with trees colonized by other bark beetles. This study explores
the spatial and temporal relationships between mountain pine beetle and non-eruptive bark
beetle species in lodgepole pine stands of southern British Columbia. Increasing populations
of non-eruptive bark beetles were positively correlated with each other, and with endemic
mountain pine beetle. Endemic and incipient-epidemic levels of mountain pine beetle were
often positively spatially associated with the bark beetles Pseudips mexicanus, Orthotomicus
latidens, Ips pini, and Hylurgops species, which themselves frequently colonized the same
host trees. As populations grew, mountain pine beetle shifted from attacking
injured/previously colonized hosts to uncolonized hosts. Identifying these potential triggers
of population phase transitions may help prevent future epidemics in areas of economic
importance.

n

TABLE OF CONTENTS
Page
ii

ABSTRACT.
TABLE OF CONTENTS

iii

LIST OF TABLES

v

LIST OF FIGURES

viii

ACKNOWLEDGEMENTS

x

1: GENERAL INTRODUCTION
l.l. Literature cited

1
8

2: TEMPORAL ASSOCIATIONS BETWEEN DENDROCTONUS
PONDEROSAE AND NON-ERUPTIVE SPECIES OF BARK BEETLES IN
STANDS OF LODGEPOLE PINE IN SOUTHERN BRITISH COLUMBIA
2.1 Abstract
2.2 Introduction
2.3 Methods
2.3.1 Study sites
2.3.2 The bole-infesting bark beetle assemblage
2.3.3 Temporal analyses
2.4 Results
2.5 Discussion
2.6 Literature cited

16
16
17
20
20
21
23
26
32
37

3. SPATIAL ASSOCIATIONS OF MOUNTAIN PINE BEETLE,
DENDROCTONUS PONDEROSAE, WITH SECONDARY BARK BEETLES
IN THE ENDEMIC TO INCIPIENT-EPIDEMIC PHASE TRANSITION
3.1 Abstract
3.2 Introduction
3.3 Methods
3.4 Results
3.5 Discussion
3.6 Literature cited

44
44
45
48
50
63
69

4. SPATIAL ASSOCIATIONS OF THE SECONDARY BARK BEETLES
DENDROCTONUS MURRAYANAE, HYLURGOPS SPR, IPS PINI,
ORTHOTOMICUS LATIDENS AND PSEUDIPS MEXICANUS IN
LODGEPOLE PINE STANDS OF SOUTHERN BRITISH COLUMBIA
4.1 Abstract

77
77

in

4.2 Introduction
4.3 Methods
4.4 Results
4.5 Discussion
4.6 Literature cited

Page
78
80
82
95
99

5. GENERAL CONCLUSIONS
5.1 Literature cited

104
108

APPENDICES
APPENDIX A
APPENDIX B
APPENDIX C
APPENDIX D
APPENDIX E: R code for analysing statistical point process models
APPENDIX F: Study of the effect of Pseudips mexicanus on host selection
behaviour of Dendroctonus ponderosae in cut bolts using no-choice
bioassays in the laboratory

IV

110
Ill
114
127
145

147

LIST OF TABLES
Page
Table 2.1:

Criteria used to estimate the number of years since (A) partial, or (B)
complete attacks by bole-infesting bark beetles on lodgepole pine trees
within seven stands at two sites in southern British Columbia between 1999
and 2002.

25

Table 3.1: Number of lodgepole pine trees strip-attacked by mountain pine beetle.
Within those trees, the numbers bearing injuries and/or any one or multiple
colonizations of other bark beetles {Dendroctonus murrayanae, Hylurgops
spp., Ips pini, Orthotomicus latidens, Pseudips mexicanus) are listed.

52

Table 3.2: Number of lodgepole pine trees mass-attacked by mountain pine beetle.
Within those trees, the numbers bearing injuries and/or any one or multiple
colonizations of other bark beetles {Dendroctonus murrayanae, Hylurgops
spp., Ips pini, Orthotomicus latidens, Pseudips mexicanus) are listed.

53

Table 3.3: Association of trees colonized by other bark beetles on the locations of trees
strip-attacked by mountain pine beetle from 2000 to 2003 in a lodgepole pine
stand of southern British Columbia (Stand B). The line in bold represents an
intercept-only model; i.e., modeling a constant density of insects across the
stand. Subsequent lines reflect whether the location of each listed insect and
year provides inference on the location of the insect studied relative to this
constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a
between-lree scale. The response variable for each equation is log(A), where
A is the estimated density of trees colonized per square meter. For example,
the estimated density of strip attack by mountain pine beetle in 2001 in
locations where all secondaries colonized trees at a rate of 0.0005/m2 or 5
trees/ha is exp(-"46 + 387(U00001) = 0.0001 or 1 trce/ha. Significant models are
listed in order of best fit for each year.
55
Table 3.4:

Association of trees strip-attacked by mountain pine beetle on the location
of mass attacks from 2002 to 2005 in a lodgepole pine stand of southern
British Columbia (Stand B). The line in bold represents an intercept-only
model; i.e., modeling a constant density of insects across the stand.
Subsequent lines reflect whether the location of each listed insect and year
provides inference on the location of the insect studied relative to this constant
density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale.
The response variable for each equation is log(A), where A is the estimated
density of trees colonized per square meter. Significant models are listed in

v

Page
59

order of best fit for each year.
Table 4.1:

Number of lodgepole pine, Pinus contorta, trees colonized by one or more
species of secondary bark beetle (Dendroctonus murrayanae, Hylurgops
spp., Ips pini, Orthotomicus latidens, and/or Pseudips mexicanus) and, within
those same trees, the number bearing injuries within seven stands in
southern British Columbia.
85

Table 4.2: Best explanatory models for the location of Orthotomicus latidens
colonization from 2001 to 2005 in lodgepole pine of stand A in southern
British Columbia. The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent
lines reflect whether the location of each listed insect and year provides
inference on the location of the insect studied relative to this constant density.
A positive estimate for a slope reflects positive spatial association, while a
negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated
density of trees colonized per square meter. For example, the estimated
density of O. latidens colonization in 2002 in locations where Pseudips
mexicanus colonized trees at a rate of 0.0005/m2 or 5 trees/ha is
exp (971 + 2124K00005)= 0.0002 or 2 trees/ha. Significant models are listed in
order of best fit for each year.
86
Table 4.3: Best explanatory models for the location of Pseudips mexicanus
colonization from 2000 to 2005 in lodgepole pine of stand A in southern
British Columbia. The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent
lines reflect whether the location of each listed insect and year provides
inference on the location of the insect studied relative to this constant density.
A positive estimate for a slope reflects positive spatial association, while a
negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated
density of trees colonized per square meter. Significant models are listed in
order of best fit for each year.
87
Table 4.4: Best explanatory models for the location of Hylurgops spp. colonization
from 2001 to 2004 in lodgepole pine of stand A in southern British
Columbia. The line in bold represents an intercept-only model; i.e., modeling
a constant density of insects across the stand. Subsequent lines reflect
whether the location of each listed insect and year provides inference on the
location of the insect studied relative to this constant density. A positive

VI

Page
estimate for a slope reflects positive spatial association, while a negative
estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of
trees colonized per square meter. Significant models are listed in order of
best fit for each year.

88

Table 4.5: Best explanatory models for the location of Ips pint colonization from 2001
to 2003 in lodgepole pine of stand A in southern British Columbia. The line
in bold represents an intercept-only model; i.e., modeling a constant density
of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect
studied relative to this constant density. A positive estimate for a slope reflects
positive spatial association, while a negative estimate reflects spatial
inhibition at a between-tree scale. The response variable for each equation is
log(A), where A is the estimated density of trees colonized per square meter.
Significant models are listed in order of best fit for each year.
89
Table 4.6: Best explanatory models for the location of Dendroctonus murrayanae
attack from 2003 and 2005 in lodgepole pine of stand D in southern British
Columbia. The line in bold represents an intercept-only model; i.e., modeling
a constant density of insects across the stand. Subsequent lines reflect
whether the location of each listed insect and year provides inference on the
location of the insect studied relative to this constant density. A positive
estimate for a slope reflects positive spatial association, while a negative
estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit
for each year.
90

vn

LIST OF FIGURES
Page
Figure 2.1: Mean number of trees colonized by various species of bark beetles per
year as a function of cruise timing. Data reflect surveys of seven stands of
lodgepole pine in southern British Columbia between 2000 and 2005.

29

Figure 2.2: Association of the number of trees colonized by one species of bark beetle
with another for the same year and census period. Data reflect surveys of
seven stands of lodgepole pine in southern British Columbia between 2000
and 2005.

30

Figure 2.3: Association of the number of trees colonized by mountain pine beetle with
other bark beetle species lagged one census period. Data reflects surveys of
seven stands of lodgepole pine in southern British Columbia between 2000
and 2005.

31

Figure 3.1: Locations of trees strip attacked by mountain pine beetle, and colonizations by
Dendroctonus murrayanae, Hylurgops spp., Ips pini, Orthotomicus latidens,
and Pseudips mexicanus in southern British Columbia, stand A, 2002.
Colonizations by mountain pine beetle and other bark beetles comprise
approximately 0.12 and 1.3% of the 19,500 pine trees in stand A respectively. 57
Figure 3.2: Location of trees strip attacked by mountain pine beetle in 2002, and mass
attacked in 2003. Strip and mass attacks comprise approximately 0.12
and 0.33% of the 19,500 lodgepole pine trees in stand A, respectively.

60

Figure 3.3: Ripley's K estimate for trees strip attacked by mountain pine beetle in 2001
for stand A. Observed estimate is shown by the black solid line, the upper
and lower limits of the 95% confidence interval are shown by the green
and blue dashes respectively. The theoretical estimate for a point process
displaying complete spatial randomness is shown by the red dashes.
The focal distance (r) on the x-axis is represented in metres.

61

Figure 3.4: Ripley's K estimate for trees strip attacked by mountain pine beetle in 2004
for stand A. Observed estimate is shown by the black solid line, the upper
and lower limits of the 95% confidence interval are shown by the green and
blue dashes respectively. The theoretical estimate for a point process
displaying complete spatial randomness is shown by the smooth red dashes.
The focal distance (r) on the x-axis is represented in metres.

62

Vlll

Page
Figure 4.1: Locations of secondary bark beetle colonization in lodgepole pine of stand A
in southern British Columbia. A) Trees colonized by Pseudips mexicanus
and Orthotomicus latidens in 2002 B) Trees colonized by P. mexicanus in
2001 and Q. latidens in 2002 comprise approximately 0.4% of the 19,500
lodgepole pine trees in the stand respectively.
91
Figure 4.2: Locations of trees colonized by Pseudips mexicanus in lodgepole pine of
stand A in southern British Columbia. Colonizations by P. mexicanus in
2001 and 2002 comprise approximately 0.4% and 0.6% of the 19,500
lodgepole pine trees in the stand respectively.

92

Figure 4.3: Locations of secondary bark beetle colonization in lodgepole pine of stand A
in southern British Columbia. A) Trees colonized by Hylurgops spp. and
Pseudips mexicanus in 2001 comprise approximately 0.2% and 0.4% of the
19,500 logepole pine trees in the stand respectively. B) Trees colonized by
H. spp. and Orthotomicus latidens in 2002 comprise approximately 0.1% and
0.4% of the 19,500 lodgepole pine trees in the stand respectively.
93
Figure 4.4: Location of trees colonized by Pseudips mexicanus and Ips pini in lodgepole
pine of stand A in southern British Columbia. Colonizations by P. mexicanus
and /. pini in 2003 comprise approximately 0.45% and 0.07% of the 19,500
lodgepole pine trees in the stand respectively.
94

IX

ACKNOWLEDGEMENTS
I wish to start by thanking my advisor, Dr. Brian Aukema, who offered me a job the
summer after undergrad and made the experience very enjoyable by surrounding me with
wonderful colleagues. It was not long and you had me convinced to continue as a Masters
student. I very much appreciate the opportunities you have afforded me and for pushing me
to be a bit bolder. I thoroughly enjoyed your humour, trying to keep up with you on the ice,
and the stimulating conversations around the lunch table about beetles, baseball, and the
bigger things of this world. Lastly, I thank you for your open-door policy, your willingness to
talk me through the things that came up outside of academics, and introducing me to Kelly's
delicious tuna melt sandwiches.
I would also like to thank my MS committee members Drs. Allan Carroll and Lisa
Poirier. Allan, thank you, particularly for sharing your data with me, guiding me along as I
grappled with an entirely new area of science, and helping me tease apart some of the
intracies of population dynamics of mountain pine beetle. Lisa thank you for jumping
onboard with enthusiasm, earnest ecological insight, and an acute attention to detail.
Deserved thanks go also to my fellow graduate students, Honey Giroday, Fraser
McKee, Ewing Teen, and Laura Machial, with whom I shared not only an office, but the
good days, bad days, and everything in between. Honey, thanks for all your help with R and
spatial modeling. It was great working with you on Pissodes strobi. Fraser, having you
sitting nearby in the office to answer questions or bounce ideas off of sped me along
tremendously. I also learned a great deal from your hustle in the field, and maybe more
importantly how to get a truck unstuck. Ewing and Laura, you came to the party a little bit
later, but I have enjoyed getting to know both of you. Ewing, I especially appreciated your
company on the drive to and from St. Paul, Minnesota.
I am indebted to the members of the Forest Insect Research Group at UNBC for their
insight on my ideas and presentations. Dezene, your presentation wizardry is legendary, but I
also appreciated our conversations about science in the broader context. Staffan, I thank you
for your fatherly kindness and sage advice. Flow many students can claim to have hung
Lindgren funnel traps with Dr. Lindgren himself? On a lighter note, Jordie Fraser I am
pleased to be able to share with you in bringing sexy back to entomology.
I also need to thank the BC Ministry of Forests & Range Protection Branch at the
Northwest Fire Centre, Ken White, Bruce Hartley and John Seinen for their help with
acquiring research material (i.e, trees and beetles). Thank you also to my field assistants
Rurik Meunter, Kathryn Berry, Gareth Hopkins and Genny Michel, as well as John Orlowski
and Steven Storch in the IK Barber Enhanced Forestry Lab for your cheerful help with my
summer projects.
Thanks especially to my friends whose support encouraged me, and who challenged
me to grow as a person these last two years. Finally, I thank my parents who also encouraged
me. Although you claimed not to understand what I was working on, you never failed to
show me how proud you were of even the smallest accomplishments.

x

CHAPTER 1
General Introduction
The weevils (Coleoptera: Curculionidae) comprise the largest of the beetle families.
With more than 50,000 species worldwide, they are easily the largest family of all animals
(Ohsawa 2005). Nearly all curculionids feed on living or dead plants, with many specializing
on woody material. The diverse group of curculionids known as bark beetles (Curculionidae,
Scolytinae) primarily feed within the subcortical region of their host trees (Coulson 1979,
Wood 1982a). Excluding a brief period of host-seeking dispersal, these insects complete
their entire life cycle in or under the bark or within the cones of their host (Rudinsky 1962).
Bark beetles are vital components of forest ecosystems as they contribute to the breakdown and turnover of senescent, weakened, dying, and dead trees (Wood 1982a). Turnover
activity is a key component of forest succession and is essential for the perpetuation of forests
with vigorously growing trees (Mattson and Addy 1975, Lundquist 1995, Jones et al. 1997).
However, many bark beetles, as agents of ecological disturbance (Raffa and Berryman 1987),
may increasingly pose a threat to previously unsuitable habitats in concert with a changing
climate (Carroll et al. 2004, Hicke et al. 2006).
Bark beetles have been informally classified as "primary" or "secondary" species,
depending on the characterization of their colonization behaviour (Rudinsky 1962, Wood
1982a). "Primary" bark beetles are generally more aggressive species that are capable of
overcoming the defenses of healthy trees. Typically, under outbreak conditions, primary bark
beetles rely on the death of their host in order to successfully complete their life cycle

1

(Berryman 1972). Large population fluctuations of primary bark beetles tend to be
intermittent. In outbreak situations, these eruptive herbivores can cause landscape-level
mortality to mature trees (Amman 1977, Wood 1982b, Safranyik and Carroll 2006).
"Secondary" species typically reproduce in material from weakened or dying host
trees, including those damaged by fire, lightning, windthrow, drought, disease, and
defoliation, as well as those suppressed by competition (Rudinsky 1962, Wood 1982a).
Subsistence in weakened trees is not universally true, however. For example pine engraver
beetle, Ips pini (Say) may kill healthy trees on occasion when populations reach sufficient
numbers (Paine et al. 1997, Steed and Wagner 2004).
Even though records of outbreaks date back only to the early 1900's, mountain pine
beetle, Dendroctonus ponderosae Hopkins, is believed to have been inhabiting pine
ecosystems in western North America for millennia (Amman 1977, Seybold et al. 2000,
Taylor and Carroll 2004). The insect is capable of colonizing both native and exotic species
of pine within its range (Furniss and Schenk 1969, Cerezke 1995). In outbreak stages, aduit
beetles are able to overwhelm the defenses of vigorous host trees through mass attacks
mediated by pheromones (Vite and Pitman 1968, Raffa and Berryman 1983) and by
innoculating hosts with mutualistic fungi (Francke-Grossman 1967, Berryman 1972,
Safranyik et al. 1974). These fungi are transported in specialized cuticular structures called
mycangia (Paine et al. 1997). Fungi exhaust the defensive capacity of host trees, and may
also provide nutritional benefits for phloem-feeding larvae (Ayres et al. 2000, Bleiker and Six
2007).

2

Development of mountain pine beetle is temperature-dependent. Adults typically
emerge in late July through August, disperse via flight, and seek new hosts (Rasmussen 1974,
Bentz et al. 1991). Female beetles construct straight vertical galleries and lay individual eggs
within niches along the sides of these galleries (Furniss and Carolin 1977). Mountain pine
beetles typically require one year to complete their life cycle, overwintering as larvae or
adults (Furniss and Carolin 1977). Semivoltine populations may be found in areas of higher
elevations or cool summer temperatures. Bivoltinism is also possible in some areas (Bentz et
al. 2001), as low-elevation sugar pine, Pinus lambertiana, of California, may produce two
generations per year (Furniss and Carolin 1977, Amman et al. 1990).
The current outbreak of mountain pine beetle within the Canadian provinces of
British Columbia and Alberta as well as the northwestern United States has overwhelmed an
unprecedented number of pine hosts. The British Columbia Ministry of Forests has reported
that the cumulative area of attacked trees within British Columbia alone extends over 16.3
million hectares, comprising the vast majority of the mature lodgepole pine, Pinus contorta
Douglas ex Louden, in the province (Westfall and Ebata 2009). Most disconcerting are the
facts that the beetle has breached the historic Rocky Mountain geoclimatic barrier (Robertson
et al. 2009, de la Giroday et al. 2010), is capable of reproducing in jack pine, Pinus
banksiana Lamb. (Furniss and Schenk 1969, Cerezke 1995), and, with increased climatic
suitability, poses a threat to Canada's boreal forest (Nealis and Peter 2008, Safranyik et al.
2010).
The economic impact of outbreaks by mountain pine beetle has fostered extensive

3

research focusing on the epidemic phase of the insect (Amman 1972, Safranyik et al. 1974,
Berryman 1976, Thomson and Shrimpton 1984, Logan et al. 1998). A great deal is known
about the biology of mountain pine beetle (Lyon 1958, Lanier and Wood 1968, Furniss and
Carolin 1977, Safranyik 1988, Bentz et al. 1991, Pureswaran and Borden 2003, Safranyik and
Carroll 2006, Safranyik et al. 2010) and its host selection behaviour at epidemic levels (Cole
and Amman 1969, Geiszler et al. 1980, Hynum and Berryman 1980, Moeck et al. 1981, Raffa
and Berryman 1982, Moeck and Simmons 1991, Pureswaran and Borden 2005). However,
beetle populations are typically found at very low levels in endemic phases. Endemic
populations are found in isolated pockets across the landscape and have been defined to
consist of approximately forty beetles per hectare (Carroll et al. 2006). Consequently, the
amount of research pertaining to the endemic phase is severely limited, as finding endemic
beetles often poses a significant challenge (Tkacz and Schmitz 1986, Bartos and Schmitz
1998, Carroll etal. 2006).
Drawing on the knowledge of the behaviour of epidemic mountain pine beetle, it
seems unlikely that endemic level beetles would be capable of successfully attacking
vigorous hosts since their population densities do not reach the numbers required for mass
attack (Raffa and Berryman 1983). Endemic populations must, therefore, be restricted to
weakened and dying host trees that are unable to mount a sustained defensive response
against colonization. Consequently, we might expect endemic levels of mountain pine beetle
to behave much like secondary bark beetles. Secondary bark beetles may interact
competitively with endemic mountain pine beetle and inhibit its reproductive success, or,

4

alternatively, may facilitate the persistence of endemic populations (Carroll et al. 2006).
A great deal is known about secondary bark beetles, their pheromones, and their
interactions with other bark beetles and/or predators, particularly species from the Ips genus
(e.g., Miller and Borden 1985, Miller et al. 1991, Miller and Borden 1992, Seybold et al.
1995, Poland and Borden 1998, Savoie et al. 1998, Aukema and Raffa 2000, Pureswaran et al.
2000, Erbilgin et al. 2002, Aukema et al. 2004). Secondary bark beetles may limit the
reproductive success of mountain pine beetle at epidemic levels through interspecific
competition (Bergvinson and Borden 1991, Rankin and Borden 1991, Safranyik et al. 1999,
Boone et al. 2008) or they may partition host resources to limit competition as has been
proposed in other bark beetle systems (Paine et al. 1981, Wagner et al. 1985, Byers 1989,
Raffa 1991, Schlyter and Anderbrant 1993, Ayres et al. 2001). In either case, the interactions
between endemic populations of mountain pine beetle and other bark beetles have not been
well studied (Carroll et al. 2006, Smith et al. 2009). Thus, I examined the population
dynamics and spatial interactions of the primary bark beetle mountain pine beetle with a
number of secondary bark beetles including Pseudips mexicamts (Hopkins), Orlhotomicus
latidens (LeConte), Hylurgops porosus (LeConte), H. rugipennis (Mannerheim), l.pini, and
D. murrayanae (Hopkins) in lodgepole pine stands undergoing population eruptions of
mountain pine beetle from the endemic to the incipient-epidemic phase.
Ecological interactions can be assessed on a series of spatial and temporal scales,
depending on the system. In forest systems, ecological interactions may occur at the tree,
stand, and landscape levels, over a broad spectrum of time scales from days to decades.

5

Coulson (1979) and White and Powell (1997) identified some challenges of studying bark
beetles, such as defining a framework for studies of population dynamics. This research
focused primarily on stand-level interactions over the course of several years.
In my thesis, I primarily used data collected from a study pioneered by Allan Carroll,
formerly with the Canadian Forest Service of Natural Resources Canada. In this study,
endemic to incipient-epidemic level phase transitions of populations of mountain pine beetle
were monitored in seven lodgepole pine stands over the course of five years. In the first data
chapter (Chapter 2), I examined the temporal interactions between secondary bark beetle
species and mountain pine beetle. In the second data chapter (Chapter 3), I explored the
interactions of secondary bark beetles with mountain pine beetle and looked at the growth of
incipient populations of mountain pine beetle. The third data chapter (Chapter 4), examined
the spatial relationships between secondary bark beetles and their associations with vigourimpaired trees during the endemic to incipient-epidemic phase transition of mountain pine
beetle. General conclusions explore the significance of the results with respect to
implications for prospective management. The appendices contain a summary of stand
characteristics and colonization by bark beetles of the bole-infesting assemblage, as well as
supplementary data for each chapter. The final appendix contains a laboratory bioassay that
explored whether endemic populations of mountain pine beetle preferentially select trees
colonized by P. mexicanus. Starved beetles were used as surrogates for endemic insects.
This thesis was written in a format where each chapter, though interrelated, is meant
to be a stand-alone entity that will be disseminated to a peer reviewed journal upon

6

successful thesis defense. As such, a small degree of ovlerlap may occur across chapters,
especially in providing research context, in order to maintain chapter integrity.

LITERATURE CITED
Amman, G. D. 1972. Mountain pine beetle brood production in relation to thickness of
lodgepole pine phloem. Journal of Economic Entomology, 65: 138-140.
Amman, G. D. 1977. The role of mountain pine beetle in lodgepole pine ecosystems
impact on succession, pp. 3-18. In W. J. Mattson (ed.), Proceedings in life sciences
the role of arthropods in forest ecosystems. Springer-Verlag, New York, NY.
Amman, G. D., M. D. McGregor, and R. E. Dolph Jr. 1990. Mountain pine beetle (FIDL).
Forest Insect & Disease Leaflet 2. United States Department of Agriculture, Ogden,
UT.
Aukema, B. H., and K. F. Raffa. 2000. Chemically mediated predator-free space: herbivores
can synergize intraspecific communication without increasing risk of predation.
Journal of Chemical Ecology, 26: 1923-1939.
Aukema, B. FL, G. R. Richards, S. J. Krauth, and K. F. Raffa. 2004. Species assemblage
arriving at and emerging from trees colonized by Ips pint in the great lakes region:
partitioning by time since colonization, season, and host species. Annals of the
Entomological Society of America. 97: 117-129.
Ayres, B. D., M. R Ayres, M. D. Abrahamson, and S. A Teale. 2001. Resource partitioning
and overlap in three sympatric species of Ips bark beetles (Coleoptera: Scolytidae).
Oecologia, 128: 443-453.
Ayres, M. P., R. T. Wilkens, J. J. Ruel, M. J. Lombardero, and E. Vallery. 2000. Nitrogen
budgets of phloem-feeding bark beetles with and without symbiotic fungi. Ecology,
81: 2198-2210.
Bartos, D. L., and R. F. Schmitz. 1998. Characteristics of endemic level mountain pine beetle
populations in south central Wyoming. Research Paper RMRS-RP-13. United States
Department of Agriculture, Forest Service, Rocky Mountain Research Station, Ogden,
UT.
Bentz, B. J., J. A. Logan, and G. D. Amman. 1991. Temperature-dependent development of
the mountain pine beetle (Coleoptera: Scolytidae) and simulation of its phenology.
The Canadian Entomologist, 123: 1083-1094.
Bentz, B. J., J. A. Logan, and J. C. Vandygriff. 2001. Latitudinal variation in Dendroctonous
pondewsae (Coleoptera: Scolytidae) development time and adult size. The Canadian
Entomologist, 133: 375-387.
8

Bergvinson, D. J., and J. H. Borden. 1991. Glyphosate-induced changes in the attack success
and development of the mountain pine beetle and impact of its natural enemies.
Entomologia Experimentalis et Applicata, 60: 203-212.
Berryman, A. A. 1972. Resistance of conifers to invasion by bark beetle-fungus associations.
Bioscience, 22: 598-602.
Berryman, A. A. 1976. Theoretical explanation of mountain pine beetle dynamics in
lodgepole pine forests. Environmental Entomology, 5: 1225-1233.
Bleiker, K. P., and D. L. Six. 2007. Dietary benefits of fungal associates to an eruptive
herbivore: potential implications of multiple associates on host population dynamics.
Environmental Entomology, 36: 1384-1396.
Boone, C. K., D. L. Six, and K. F. Raffa. 2008. The enemy of my enemy is still my enemy:
competitors add to predator load of a tree-killing bark beetle. Agricultural and Forest
Entomology, 10:411-421.
Byers, J. A. 1989. Chemical ecology of bark beetles. Experientia, 45: 271-283.
Carroll, A. L., B. H. Aukema, K. F. Raffa, G. D. Smith, and B. S. Lindgren. 2006. Mountain
pine beetle outbreak development: the endemic - incipient transition. Mountain Pine
Beetle Initiative, Project 1.03 Working Paper, Natural Resources Canada, Canadian
Forest Service. 21p.
Carroll, A. L., S. W. Taylor, J. Regniere, and L. Safranyik. 2004. Effects of climate change on
range expansion by the mountain pine beetle in British Columbia, pp. 223-232. In
T. L. Shore. J. E. Brooks, and J. E. Stone (eds.), Mountain pine beetle symposium:
challenges and solutions. Natural Resources Canada, Canadian Forest Service, Pacific
Forestry Centre, Information Report BC-X-399, Victoria, British Columbia, Canada.
Cerezke, H. F. 1995. Egg gallery, brood production, and adult characteristics of mountain
pine beetle, Dendroctonus ponderosae Hopkins (Coleoptera: Scolytidae), in three pine
hosts. The Canadian Entomologist, 127: 955-965.
Cole, W. E., and G. D. Amman. 1969. Mountain pine beetle infestations in relation to
lodgepole pine diameters. Res. Note INT-95. United States Department of
Agriculture, Forest Service, Intermountain Forest and Range Experiment Station,
Ogden, UT.
Coulson, R. N. 1979. Population dynamics of bark beetles. Annual Review of Entomology,
24: 417-447.
9

de la Giroday, H.-M., A. L. Carroll, and B.H. Aukema. 2010. Breach of the northern Rocky
Mountain geoclimatic barrier: Initiation of range expansion by the mountain pine
beetle. Presubmission review.
Erbilgin, N. E., V. Nordheim, B. H. Aukema, and K. F. Raffa. 2002. Population dynamics of
Ips pini and Ips grandicollis in red pine plantations in Wisconsin: within- and
between-year associations with predators, competitors, and habitat quality.
Environmental Entomology, 31: 1043-1051.
Francke-Grossman, H. 1967. Ectosymbiosis in wood-inhabiting insects, pp. 141-203. In S.
M. Henry (ed.), Symbiosis. Academic Press, New York, NY.
Furniss, M. M., and J. A. Schenk. 1969. Sustained natural infestation by the mountain pine
beetle in seven new Pinus and Picea hosts. Journal of Economic Entomology, 62:
518-519.
Furniss, R. L., and V. M. Carolin. 1977. Western Forest Insects. United States Forest Service
Miscellaneous Publication 1339. United States Department of Agriculture, Forest
Service, Washington, D.C.
Geiszler, D. R., V. F. Gallucci, and R. I. Gara. 1980. Modeling the dynamics of mountain
pine beetle aggregation in a lodgepole pine stand. Oecologia, 46: 244-253.
Hicke, J. A., J. A. Logan, J. Powell, and D. S. Ojima. 2006. Changing temperatures influence
suitability for modeled mountain pine beetle (Dendroctonus pondewsae) outbreaks in
the western United States. Journal of Geophysical Research, 111: G02019,
doi:l0.1029/2005JG000101.
Hynum, B. G., and A. A. Bcrryman. 1980. Dendroctonus ponderosae (Coleoptera:
Scolytidae): pro-aggregation landing and gallery initiation on lodgepole pine. The
Canadian Entomologist, 112: 185-191.
Jones, C. G., J. H. Lawton, and M. Shachak. 1997. Positive and negative effects of organisms
as physical ecosystem engineers. Ecology, 78: 1946-1957.
Lanier, G. N., and D. L. Wood. 1968. Controlled mating, karyology, morphology, and sexratio in the. Dendroctonus ponderosae complex. Annals of the Entomological Society
of America, 61: 517-526.
Logan, J. A., P. White, B. A. Bentz, and J. A. Powell. 1998. Model analysis of spatial patterns
in mountain pine beetle outbreaks. Theoretical Population Biology, 53: 236-255.

10

Lundquist, J. E. 1995. Pest interactions and canopy gaps in ponderosa pine stands in the
Black Hills, South Dakota, USA. Forest Ecology and Management, 74: 37-48.
Lyon, R. L. 1958. A useful secondary sex character in Dendroctonus bark beetles. The
Canadian Entomologist, 90: 582-584.
Mattson, N. D., and W. J. Addy. 1975. Phytophagous insects as regulators of forest primary
production. Science, 190: 515-522 .
Miller, D. R., and J. H. Borden. 1985. Life history and biology of Ips latidens (Leconte)
(Coleoptera: Scolytidae). The Canadian Entomologist, 117: 859-871.
Miller, D. R., and J. H. Borden. 1992. (S)-(+)-Ipsdienol: interspecific inhibition of Ips
latidens (Leconte) by Ips pini (Say) (Coleoptera: Scolytidae). Journal of Chemical
Ecology, 18: 1577-1582.
Miller, D. R., J. H. Borden, G. G. S. King, and K. N. Slessor. 1991. Ipsenol: an aggregation
pheromone for Ips latidens (LeConte) (Coleoptera: Scolytidae). Journal of Chemical
Ecology, 17: 1517-1527.
Moeck, H. A., and C. S. Simmons. 1991. Primary attraction of mountain pine beetle,
Dendroctonus ponderosae Hopkins (Coleoptera: Scolytidae), to bolts of lodgepole
pine. The Canadian Entomologist, 123: 299-304.
Moeck, H. A., D. L. Wood, and K. Q. Lindahl Jr. 1981. Host selection behaviour of bark
beetles (coleoptera: scolytidae) attacking Pinus ponderosa, with special emphasis on
the western pine beetle, Dendroctonus brevicomis. Journal of Chemical Ecology, 7:
49-83.
Nealis, V. G., and B. Peter. 2008. Risk assessment of the threat of mountain pine beetle to
Canada's boreal and eastern pine resources. Information Report BC-X-417, Pacific
Forestry Centre, Canadian Forest Service, Natural Resources Canada, Victoria,
British Columbia. 38p.
Ohsawa, M. 2005. Species richness and composition of Curculionidae (Coleoptera) in a
conifer plantation, secondary forest, and old-growth forest in the central mountainous
region of Japan. Ecological Research, 20: 632-645.
Paine, T. D., M. C. Birch, and P. Svihra. 1981. Niche breadth and resource partioning by four
sympatric species of bark beetles (Coleoptera: Scolytidae). Oecologia, 48: 1-6.

1!

Paine, T. D., K. F. Raffa and T. C. Harrington. 1997. Interactions among scolytid bark beetles,
their associated fungi, and live host conifers. Annual Review of Entomology, 42:
179-206.
Poland, T. M., and J. H. Borden. 1998. Competitive exclusion of Dendroctonus rufipennis
induced by pheromones of Ips tridens and Dryocoetes affaber (Coleoptera:
Scolytidae). Journal of Economic Entomology, 91: 1150-1161.
Pureswaran, D. S., and J. H. Borden. 2003. Is bigger better? Size and pheromone production
in the mountain pine beetle, Dendroctonus ponderosae Hopkins (Coleoptera:
Scolytidae). Journal of Insect Behavior, 16: 765-782.
Pureswaran, D. S., and J. H. Borden. 2005. Primary attraction and kairomonal host
discrimination in three species of Dendroctonus (Coleoptera: Scolytidae).
Agricultural and Forest Entomology, 7: 219-230.
Pureswaran, D. S., R. Gries, J. H. Borden, and H. D. Pierce, Jr. 2000. Dynamics of
pheromone production and communication in the mountain pine beetle,
Dendroctonus ponderosae Hopkins, and the pine engraver, Ips pint (Say) (Coleoptera:
Scolytidae). Chemoecology, 10: 153-168.
Raffa, K. F. 1991. Temporal and spatial disparities among bark beetles, predators, and
associates responding to synthetic bark beetle pheromones: Ips pint (Coleoptera:
Scolytidae) in Wisconsin. Environmental Entomology, 20: 1665-1679.
Raffa, K. F., and A. A. Berryman. 1982. Gustatory cues in the orientation of Dendroctonus
ponderosae (Coleoptera: Scolytidae) to host trees. The Canadian Entomologist, 114:
97-104.
Raffa, K. F., and A. A. Berryman. 1983. The role of host plant resistance in the colonization
behavior and ecology of bark beetles (Coleoptera: Scolytidae). Ecological
Monographs, 53: 27-49.
Raffa, K. F. and A. A. Berryman. 1987. Interacting selective pressures in conifer-bark beetle
systems: a basis for reciprocal adaptations? The American Naturalist, 129: 234-262.
Rankin, L. J., and J. H. Borden. 1991. Competitive interactions between the mountain pine
beetle and the pine engraver in lodgepole pine. Canadian Journal of Forest Research,
21:1029-1036.

12

Rasmussen, L. A. 1974. Flight attack and behavior of mountain pine beetles in lodgepole
pine of northern Utah and southern Idaho. Research Note INT-180. United States
Department of Agriculture, Forest Service, Intermountain Research Station, Ogden,
UT.
Robertson, C , T. A. Nelson, D. E. Jelinski, M. A. Wulder, and B. Boots. 2009. Spatialtemporal analysis of species range expansion: the case of the mountain pine beetle,
Dendroctonous ponderosae. Journal of Biogeography, 36: 1446-1458.
Rudinsky, J. A. 1962. Ecology of Scolytidae. Annual Review of Entomology, 7: 327-348.
Safranyik, L. 1988. Estimating attack and brood totals and densities of the mountain pine
beetle in lodgepole pine trees. The Canadian Entomologist, 120: 323-331.
Safranyik, L., and A. L. Carroll. 2006. The biology and epidemiology of the mountain pine
beetle in lodgepole pine forests, pp. 3-66. In L. Safranyik and B. Wilson (eds.), The
mountain pine beetle: a synthesis of its biology, management and impacts on
lodgepole pine. Pacific Forestry Centre, Canadian Forest Service, Natural Resources
Canada, Victoria, British Columbia.
Safranyik, L., A. L. Carroll, J. Regniere, D. W. Langor, W. G. Riel, T. L. Shore, B. Peter, B. J.
Cooke, V. G. Nealis, and S. W. Taylor. 2010. Potential for range expansion of
mountain pine beetle into the boreal forest of North America . The Canadian
Entomologist, 142: 415-442.
Safranyik, L., T. L. Shore, D.A. Linton and L. Rankin. 1999. Effects of induced competitive
interactions with secondary bark beetle species on the establishment and survival of
mountain pine beetle broods in lodgepole pine. Information Report BC-X-384, Pacific
Forestry Centre, Canadian Forest Service, Natural Resources Canada, Victoria,
British Columbia. 40p.
Safranyik, L., D. M. Shrimpton, and H. S. Whitney. 1974. Management of lodgepole pine to
reduce losses from the mountain pine beetle. Forestry Technical Report 1. Pacific
Forestry Centre, Canadian Forest Service, Natural Resources Canada, Victoria,
British Columbia. 29p.
Savoie, A., J. H. Borden, H. D. Pierce Jr., R. Gries, and G. Gries. 1998. Aggregation
pheromone of Pityogenes knechteli and semiochemical-based interactions with three
other bark beetles. Journal of Chemical Ecology, 24: 321-337.
Schlyter, F., and O. Anderbrant. 1993. Competition and niche separation between two bark
beetles: existence and mechanisms. Oikos, 68: 437-447.

13

Seybold, S. J., J. Bohlman, and K. F. Raffa. 2000. Biosynthesis of coniferophagous bark
beetle pheromones and conifer isoprenoids: evolutionary perspective and synthesis.
The Canadian Entomologist, 132: 697-753.
Seybold, S. J., T. Ohtsuka, D. L. Wood, and I. Kubo. 1995. Enantiomeric composition of
ipsdienol: A chemotaxonomic character for north american populations of Ips spp.
in the pini subgeneneric group (Coleoptera: Scolytidae). Journal of Chemical
Ecology, 21:995-1016.
Smith, G. D., A. L. Carroll, and B. S. Lindgren. 2009. Life history of a secondary bark
beetle, Pseudips mexicanus (Coleoptera: Curculionidae: Scolytinae), in lodgepole
pine in British Columbia. The Canadian Entomologist, 141: 56-69.
Steed, B. R., and M. R. Wagner. 2004. Importance of log size on host selection and
reproductive success of Ips pini (Coleoptera: Scolytidae) in ponderosa pine slash of
northern Arizona and western Montana. Journal of Economic Entomology, 97:
436-450.
Taylor, S. W., and A. L. Carroll. 2004. Disturbance, forest age, and mountain pine beetle
outbreak dynamics in BC: A historical perspective, pp. 41-51. In T. L. Shore, J. E.
Brooks, and J. E. (eds.), Mountain Pine Beetle Symposium: Challenges and
Solutions. October 30-31, 2003, Kelowna, British Columbia. Natural Resources
Canada, Canadian Forest Service, Pacific Forestry Centre, Information Report BCX-399, Victoria, BC.
Thomson, A. J., and D. M. Shrimpton. 1984. Weather associated with the start of mountain
pine beetle Dendroctonus ponderosae outbreaks. Canadian Journal of Forest
Research, 14: 255-258.
Tkacz, B. M., and R. F. Schmitz. 1986. Association of an endemic mountain pine beetle
population with lodgepole pine infected by armillaria root disease in Utah. Research
Note INT-353. United States Department of Agriculture, Forest Service,
Intermountain Research Station, Ogden, UT.
Vite, J. A., and G. B. Pitman. 1968. Bark beetle aggregation: effects of feeding on the release
of pheromones in Dendroctonus and Ips. Nature, 218: 160-170.
Wagner, T. L., R. O. Flamm, and R. N. Coulson. 1985. Strategies for cohabitation among the
southern pine bark beetle species: comparisons of life-process biologies. General
Technical Report SO-56. United States Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station, Asheville, NC.

14

Westfall, J., and T. Ebata. 2009. 2008 summary of forest health conditions in British
Columbia. Ministry of Forests and Range, Forest Practices Branch. British Columbia
Forest Service, Victoria, British Columbia, Canada.
White, P., and J. Powell. 1997. Phase transition from environmental to dynamic determinism
in mountain pine beetle attack. Bulletin of Mathematical Biology, 59: 609-643.
Wood. S. L. 1982a. The role of pheromones, kairomones, and allomones in the host selection
and colonization behavior of bark beetles. Annual Review of Entomology, 27:
411-446.
Wood. S. L. 1982b. Editor. The bark and ambrosia beetles of North and Central America
(Coleoptera: Scolytidae), a taxonomic monograph. Brigham Young University,
Provo, Utah. 1359p.

15

CHAPTER 2
Temporal associations between Dendroctonus ponderosae and non-eruptive species of bark
beetles in stands of lodgepole pine in southern British Columbia
ABSTRACT
The majority of our knowledge of the ecology of mountain pine beetle, Dendroctonus
ponderosae Hopkins, originates from studies of epidemic-level populations. Less is known
about what factors might trigger population transitions from endemic to incipient-epidemic
levels. The population dynamics of mountain pine beetle may be influenced by associations
with trees colonized by secondary bark beetles, particularly when the former is at endemic
levels and existing in habitat colonized by the latter. Temporal relationships between
mountain pine beetle and species of secondary bark beetles comprising part of the boleinfesting bark beetle assemblage were examined over five years in seven lodgepole pine
stands of southern British Columbia where mountain pine beetle was erupting from endemic
to epidemic levels. Prior to the transition of populations of mountain pine beetle from
endemic to incipient-epidemic levels, the number of trees attacked by secondary bark beetles
increased. Increasing populations of secondaries were positively correlated with each other,
and with increasing populations of endemic mountain pine beetle in all stands. Identifying
potential triggers of population phase transitions may enable the minimization of mountain
pine beetle epidemics in areas of economic, cultural, aesthetic, and/or recreational
importance.
Key words: population dynamics; temporal dependence; interspecific competition

16

INTRODUCTION
Populations of phytophagous insects rise and fall under the influence of endogenous
and exogenous pressures, and may exhibit points of stable equilibrium, cyclic oscillations, or
a lack of periodicity (May 1974). Periodic oscillations, for example, are often seen in
lepidopteran defoliator systems such as gypsy moth, Lymantria dispar (L.) (Williams and
Liebhold 1995, Johnson et al. 2005), larch budmoth, Zeiraphera diniana (Guenee)
(Baltensweiler and Fischlin 1988), spruce budworm, Choristoneura fumiferana (Clemens)
(Blais 1965), and forest tent caterpillar, Malacosoma disstria (HiAbner) (Cooke and Lorenzetti
2006). Aperiodical population fluctuations are often found in tree-killing bark beetle
systems, and generally occur in an eruptive manner. Primary examples include southern pine
beetle, Dendroctonus frontalis (Zimmerman), spruce bark beetle, Ips typographus (L.)
(0kland and Bj0rnstad 2006), and mountain pine beetle, D. ponderosae Hopkins (Raffa et al.
2008), a species of particular relevance to this study.
Mountain pine beetle is an eruptive species of bark beetle with a broad geographic
range stretching across much of western North America (Safranyik and Carroll 2006). It is
capable of colonizing nearly every species of native and introduced pine within this range
(Furniss and Schenk 1969, Smith et al. 1981, Cerezke 1995, Carroll et al. 2004). British
Columbia is currently experiencing the largest outbreak of mountain pine beetle in recorded
history (Westfall and Ebata 2009). At outbreak levels, mass attacks coordinated by
pheromones, in concert with vectored fungi, enable the beetles to overwhelm the defenses of
healthy, large-diameter trees (Wood 1982a, Raffa and Bcrryman 1983). Mass attacks are

17

promoted by synchronous emergence of adults between late July and early August, achieved
through temperature-dependent development (Rasmussen 1974, Bentz et al. 1991, Safranyik
et al. 2010). Outbreaks generally begin to decline with the depletion of mature hosts capable
of sustaining an epidemic, sometimes in concert with mortality of brood caused by cold
winter temperatures and larval desiccation (Reid 1963, Cole and Amman 1969, Safranyik et
al. 1974, Amman 1984, Safranyik and Linton 1998, Regniere and Bentz 2007, Sambaraju et
al. 2011). As mountain pine beetle populations begin to decline, they may also be
outcompeted by other species such as I. pint (Say) whose populations build up in the tops of
hosts killed by mountain pine beetle (Furniss and Carolin 1977, Rankin and Borden 1991).
Although the decline of outbreaks of mountain pine beetle is fairly well understood,
there are still questions surrounding the growth of endemic populations. Mountain pine
beetle populations typically exist for long periods at endemic phases. Researchers have long
puzzled over what triggers an outbreak and why populations may persist at endemic levels in
one area, but erupt in another (Logan et al. 1998), or erupt simultaneously over large areas
(Aukema et al. 2006). Despite best efforts, hazard rating systems frequently fail to predict the
risk of tree mortality by mountain pine beetle (Bentz 1993, Nelson ct al. 2008). Favourable
conditions for beetle reproduction include successive warm summers, mild winters, and stress
events such as drought (Reid 1963, Safranyik et al. 1974). It is believed that when such
conditions coincide with an adequate number of mature hosts, mountain pine beetle may
enter the incipient-epidemic phase (Carroll et al. 2006). Because endemic mountain pine
beetle is likely limited by its inability to colonize healthy hosts (Raffa and Berryman 1983), a

18

theory of facilitation between "secondary" bark beetles and this primary bark beetle is
beginning to emerge (Carroll et al. 2006).
"Secondary" species of bark beetles reproduce in the phloem of weakened, dead, and
dying trees (Wood 1982a). Most species of secondary bark beetles remain at relatively low
population levels, and contribute to the break-down and turnover of senescent and dead trees
(Wood 1982a). Secondary species of bark beetles that may share hosts with mountain pine
beetle in British Columbia include D. murrayanae (Hopkins), Hylurgops porosus (LeConte),
H. rugipennis (Mannerheim), /. pini (Say), Orthotomicus latideas (LeConte), and Pseudips
inexicanus (Hopkins). Many of these species are multivoltine, and, although their emergence
and flight periods vary considerably, generally precede mountain pine beetle's flight in late
summer (Schenk and Benjamin 1969, Miller and Borden 1985, Safranyik et al. 2000,
Safranyik et al. 2004, Safranyik and Carroll 2006, Furniss and Kegley 2008, Smith et al.
2009).
This study examined the temporal interactions between endemic level mountain pine
beetle and several species of secondary bark beetles as populations begin to build toward an
epidemic. The current study explored five years of bark beetle colonization within seven
lodgepole pine stands in two sites in southern British Columbia immediately prior to a
population eruption of mountain pine beetle. The population dynamics of endemic mountain
pine beetle within each stand were examined to identify any temporal associations with
secondary bark beetle species during the transition from the endemic to the incipientepidemic phase. Temporal associations between mountain pine beetle and other bole-

19

infesting bark beetles may provide inference regarding the mechanism of this phase transition
and establish context for further spatial analyses in which these insects persist through time
by partitioning host resources within and between trees.
METHODS
Study Sites
Two sites were established within southern British Columbia per Carroll et al. (2006).
In brief, the site selection criteria included a historically suitable climate for mountain pine
beetle, and a lack of tree-killing activity by the insect within 10 km of the sites. The first site,
located at Angstad Creek, 25 km south of Merritt, was established in 2002. The second site,
located on the Aberdeen Plateau, 35 km northeast of Kelowna, was established in 2003.
At Angstad Creek, three lodgepole pine stands were initially identified for study
(stands A, B, and C). On the Aberdeen Plateau, two stands were chosen (stands D and E).
An additional stand at Angstad Creek, and Aberdeen Plateau (stands E and G, respectively),
were later added to the study to replace those stands in which mountain pine beetle
populations transitioned from the endemic to the incipient-epidemic phase.
Stands were chosen to represent optimal mountain pine beetle habitat, i.e., lodgepole
pine-leading, greater than 80 years old, and moderately dense (800 - 1500 stems/ha)
(Safranyik and Carroll 2006). Furthermore, only stands with distinct boundaries formed by
topographical features (e.g., water bodies, roads, clear cuts) or ecological conditions (e.g.,
forest age or species changes) were selected. These criteria were established to minimize the
potential effects of immigration and emigration associated with immediately adjacent

20

habitats, and thereby ensure assessments of local population dynamics. Stands chosen for
study at each site were at least 1 km apart.
Following stand selection, variable radius prism plots were established within each
stand at a density of one plot per hectare to ascertain average stand mensurational
characteristics using the methods of Avery and Burkhart (2002). Tree diameter was
measured at breast height (1.3 m). Tree height was determined using a laser hypsometer, and
tree age was ascertained from cores collected at breast height.
The states of all trees in each variable radius plot were also assessed for conditions
that could potentially impair tree vigour. These conditions included mechanical damage to
the main stem or roots, competitive status (suppressed versus dominant), root or foliar
infections, and previous non-lethal infestations by herbivorous insects.
The bole-infesting bark beetle assemblage
For purposes of spatial and temporal characterization of the bole-infesting bark beetle
assemblage, a 25 x 50 m reference grid system was generated within each stand. An initial
census was conducted to establish a baseline of all previous activity by bark beetles. The
stems of all trees in each stand were carefully assessed for evidence of attack by bark beetles.
Assessments were restricted to the lower 3 m of the boles where mountain pine beetle is most
prevalent (Safranyik and Carroll 2006). The presence of boring dust in bark crevices,
defensive resin exudate, and discoloured foliage was used to ascertain potential infestations.
Portions of the bark were carefully removed in the vicinity of beetle activity (as evidenced by
entrance holes and boring dust) and species were identified either directly when individuals

21

were present or indirectly based upon diagnostic gallery patterns (Bright 1976, Wood 1982b).
When beetles were present, only sufficient bark was removed (<225 cnrVtree, ca. 15 x 15 cm
patch) to confirm the identity of the attacking species and thereby minimize impacts to their
broods.
For attacked trees in which some or all beetles had completed development and
dispersed, the year of attack was estimated based on the condition of remaining bark and
phloem, the presence of wood boring beetles and saproxylic insects that follow bark beetle
attacks, and the condition of foliage remaining on trees (Table 2.1). The accuracy of these
estimations was later confirmed through comparison to the detailed sampling described
below. Estimations of the year of attack were considered reliable for trees infested up to a
maximum of two years in the past. The height, diameter and injury condition of attacked
trees was determined as described above, and each tree was spatially referenced by recording
its distance and azimuth to the nearest grid point.
Following the initial baseline censuses, detailed assessments were conducted to
quantify variation in the abundance and distribution of the resident bole-infesting bark beetle
assemblage within and among seasons. All trees in each stand were carefully inspected at 4week intervals from early June to early September of each year (2002 - 2005 for Angstad
Creek, 2003 - 2005 for Aberdeen Plateau). Trees were assessed, marked, spatially
referenced, and their characteristics and condition recorded as described above. Due to the
demanding effort required to carefully inspect all trees in each stand at 4-week intervals, the
detailed assessments were restricted to 2 stands per site each year (initially stands A and B at

22

Angstad, D and E at Aberdeen). Thus, stand C at Angstad Creek was limited to a single
inspection in mid-September of each year to provide a summary of total bark beetle
abundance for the season.
Stands in which mountain pine beetle populations erupted in the course of the
investigation were omitted from the detailed 4-week assessments in the year following
eruption (stand B at Angstad, stand E at Aberdeen) and replaced with additional stands at
each site. New stands (stands F and G at Angstad and Aberdeen, respectively) were chosen,
established, censused and sampled in the same manner as described above.
Temporal Analyses
Graphical inspection revealed that the flight of mountain pine beetle was generally
later than other bark beetles (Fig. 2.1), so data were first grouped into "early" (June - midJuly) and "late" (mid-July - September) time periods. Linear mixed effects models were
used to examine associations between the numbers of trees colonized by each secondary
species and by mountain pine beetle. The numbers of trees colonized by each species were
incorporated as fixed effects. The variations between sites and stands within sites were
incorporated as random effects. To examine whether the number of attacks on pines by
mountain pine beetle across all stands was associated with secondary bark beetle activity in
an earlier period, the number of trees attacked by mountain pine beetle was regressed against
the number of trees colonized by secondaries lagged t-\. When assessing the associations
solely between secondaries, original time periods and colonizations from all years were used.
When assessing the associations between secondaries and mountain pine beetle, only the

23

years where mountain pine beetle was believed to be at endemic or early incipient-epidemic
population levels were used. These levels were defined to be less than five attacked trees per
hectare. Analyses between species were compared using Akaike's Information Criterion
(AIC) where the lowest AIC value indicated the best fitting regression model (Akaike 1973).
Response variables were transformed as necessary to satisfy assumptions of each model
including normal distribution, homogeneity of variances, and appropriate fit. Assumptions
of homogenous variance and normal distribution of errors were assessed using residual plots.
Only equations for the best models for each species are reported.

24

Table 2.1: Criteria used to estimate the number of years since (A) partial, or (B) complete
attacks by bole-infesting bark beetles on lodgepole pine trees within seven stands at two sites
in southern British Columbia between 1999 and 2002 (from Carroll et al. 2006).
Item Evaluated

Years since initial bark beetle attack
One

Two

Three (or more)

Bark beetle galleries
with emergence holes
confined to portion of
bole circumference
Current attacks by
Pseudips mexicanus,
Orthotomicus latidens
and/or Ips pini in green
phloem at margins of
strip attack
Current attacks by
Trypodenclron lineatitm
within region of partial
attack

No remaining moist
phloem, no visible
decay fungi within
region of partial attack
Current attacks by
Hylastes spp. of large
roots directly beneath
region of partial attack

Loose bark, decay fungi
visible within region of
partial attack

N/A

Saproxylic insects
beneath bark within
region of partial attack

Remnants of moist
phloem interspersed
among bark beetle
galleries with
emergence holes
Current attacks in
remnant phloem by
Pseudips mexicanus.
Ips pini, and/or
Orthotomicus latidens
Current attacks by
Trypodenclron lineatum

No remaining moist
phloem, no visible
decay fungi

Loose bark, decay fungi
visible

Current attacks by
Hylastes spp. of root
collar and laree roots

Ongoing attacks by
Hylastes spp. of root
collar and large roots

A. Partial attack"
Bark and phloem

Bark beetles

Wood borers and
saproxylic insects'3

B. Complete attack0
Bark and phloem

Bark beetles

Wood borers and
saproxylic insects

Current attacks by
wood-boring beetles
(Ce ram bye idae,
Buprestidae)

Ongoing attacks by
Hylastes spp. of large
roots directly beneath
region of partial attack

Emergence holes by
wood-boring beetles
(Cerambycidae,
Buprestidae), saproxylic
insects beneath bark
Red foliage, 20% needle
Foliaged
Fading or red foliage,
Red foliage, 60%
retention
100% needle retention
needle retention
a
One or more years of infestation confined to "strips" of the circumference of the bole, trees remain
alive.
'After Grove (2002)
L
Attacks around the entire circumference of the bole, trees dead.
''Adapted from Wulder et al. (2006)

25

RESULTS
The populations of mountain pine beetle in five of the seven stands (A, B, C, E, and
F) underwent a transition from endemic to incipient-epidemic levels, judged by the number of
trees colonized as a proxy for population density (a fair assumption in operational settings).
The number of trees attacked by mountain pine beetle increased in these five stands each year
(for mensurational characteristics see Appendix A, for colonization patterns see Appendix B:
Tables B.1-B.3, B.5 and B.6). In two of the stands (D and G) the number of trees attacked by
mountain pine beetle reached a small peak in 2004, but dropped substantially the following
year (Appendix B: Tables B.4, and B.7). These two stands did not appear to enter the
incipient-epidemic phase in the years under investigation.
Pseudips mexicanus colonized the most trees in all of the stands, followed by O.
latidens, and Hylurgops spp. (Appendix B: Tables B.1-B.7). Colonization on the lower bole
of trees by /. pini was found in all seven stands, but in much lower numbers than the
aforementioned species. Trees colonized by D. murrayanae were also found in all stands,
generally in smaller numbers than boles colonized by /. pini, with the exception of one stand
(stand D; Appendix B: Table B.4).
The timing of attack for each bark beetle species was examined by determining the
average number of trees colonized in each time period, each year, across all stands (Fig. 2.1).
Dendroctonus murrayanae, Hylurgops spp., O. latidens, and P. mexicanus colonized the most
trees earlier in the season (Fig. 2.1 A, B, C, and E). The number of trees colonized by O.
latidens and D. murrayanae in each month declined quite steadily from June peaks (Fig. 2.1 B

26

and E), while a more dramatic decrease could be seen for P. mexicanus and Hylurgops spp.
(Fig. 2.1 A and C). Although /. pini appeared to attack trees primarily in May or June, a
second peak in trees infested with /. pini was noted in the August censuses (Fig. 2.ID)
indicative of colonization occurring between July and August. Trees colonized by mountain
pine beetle were generally found later in the season, in the months of August and September
(Fig. 2.1F).
Temporal interactions between secondary bark beetles
Populations of many bole-infesting bark beetles appeared to be positively correlated.
In general, a greater number of trees colonized by any one secondary bark beetle species in a
given time period was highly correlated with a greater number of trees colonized by other
secondary species (see Fig. 2.2). For example, the number of trees colonized by P.
mexicanus was significantly positively correlated with the number of trees colonized by O.
latidens from the same year and time period (Fig. 2.2A). Likewise, the number of trees
colonized by D. murrayanae, Hylurgops spp., and /. pini were also positively correlated with
the abundance of P. mexicanus. However, ATC values were higher for these regression
models indicating that the number of trees colonized by O. latidens was the best predictor for
trees colonized by P. mexicanus (AIC a umdens = 776.49 < AIC». „mrra),anue = 900.87 < AICW spp =
918.06 < AIC/ ,„m = 953.60). Similarly, the number of O. latidens attacks, although positively
correlated with all species (some results not shown), was most significantly positively
correlated with the number of trees attacked by P. mexicanus in the same year (Fig. 2.2B).
Colonization by Hylurgops spp., D. murrayanae, and /. pini were also strongly

27

correlated with all secondary species (some results not shown). However their abundance
was best explained by the number of trees colonized by P. mexicanus (Fig. 2.2C, D, and E).
Temporal interactions between secondary bark beetles and mountain pine beetle
The number of trees attacked by mountain pine beetle was correlated with the number
of trees colonized by all species of secondary bark beetles (Fig. 2.3). As the number of trees
attacked by secondaries in a season increased, an increase in all types of attack (resistedattack, strip-attack, and mass-attack) by mountain pine beetle was also evident (Fig. 2.3).
The best predictor of the number of trees colonized by mountain pine beetle was the number
of trees colonized by O. latidens, (AIC = 199.75), followed by Hylurgops spp. (AIC =
200.46), P. mexicanus (AIC = 200.85), D. murrayanae (AIC= 204.51), and I. pini (AIC207.72). The positive trend between the numbers of trees colonized by mountain pine beetle
and those previously colonized by other bole-infesting bark beetles was only evident when
populations of mountain pine beetle were at either endemic or early incipient-epidemic
levels. Once colonization by mountain pine beetle reached later stage incipient-epidemic
levels, as judged by strip and mass attacks within the stands, their populations were no longer
correlated with the number of trees colonized by other bark beetles in the bole-infesting
assemblage (P>0.05 for all cases).

28

14

10

B

12
10

6
4
2
0

0

June

July

August

September

June

July

4.5
4

August

September

O latidens

P mexicanus
1.2

•

D

3.5
0.8

3
2.5

0.6

2
1.5

o 0.4

*

•

1

§ 0.2
3
0

0.5
0
June

July

August

September

June

July

Hylurgops spp

1.8
16

September

35

K

•

August
/ pini

30

1.4
25

1.2
1

20

0.8

15

0.6

o

0.4

zc

m

CO
CD

0.2

5

0
June

July
0

August

10
5
0 •
June

September

July

August

September

D ponderosae

murrayanae

Figure 2.1: Mean number of trees colonized by various species of bark beetles per year as a
function of cruise timing. Data reflect surveys of seven stands of lodgepole pine in southern
British Columbia between 2000 and 2005.

29

ym = 0.88+0.18*
F
=469;P<0.0001

B

1.220

1
0

1
1
1
20
40
60
No of trees colonized by O latidens
1/2

'

= 0.07+0.066x
=350;P<0.0001

1,220

i
0

i
0

1
1
1
1
20
40
60
80
No, of tress colonized by P mexicanus

145;/M).000l

r
100

D

'

1
1
1
1
20
40
80
80
No of trees colonized by P mexicanus

yt/2 = 0.I + 0.041 x
V
=257;P<0.0001

r
100

T

0

1
1
1
1
£0
40
60
80
No of trees colonized by P. inextcanus

r
100

E

* •
/•
«
•

•
••

/

••

• * •••

•
•

/

m•

f

•» mm
•» mmm^f^i

%tr •

•MMW»««I

"i
0

i

/

y *
•

•
*

•

/

•

*•

i

i

i

20
40
60
80
No of trees colonized by P. mexicanus

r
tOO

Figure 2.2: Association of the number of trees colonized by one species of bark beetle with
another for the same year and census period. Data reflect surveys of seven stands of
lodgepole pine in southern British Columbia between 2000 and 2005.
30

ym

= 1.52 + 0.065JC

F

= 37.1;P<0.0001
1 44

;y1/2= 1.48+0.13x
F
=31.6;P<0.0001

»

~i

1 44

T

r

0

20
40
SO
80
100
No of trees colonized by O latidens
1/2
y - 1 . 3 3 +0.047*
:34.8;/><0.0001
-IF
1 44

9

X

'

•

B

1
1
1
r~
10
20
30
40
No of trees colonized by Hylurgops spp

F

•

'

=23.2;P<0.0001
1 44

*

-

i

r

~\

1

1

r

20
40
60
80
1O0 120
No of trees co'onized by P raexicanus

140

0

5

10

15

i

i
0

i

i

i

i

v" 2 ^ 1.72+0.39v
l6 5;/ 3 =0.0002
F

n
2

i

i

i

No of trees colonized by D murrayanae

E

i

r~

4
6
8
10
12
No of trees colonized by I pini

Figure 2.3: Association of the number of trees colonized by mountain pine beetle with other
bark beetle species lagged one census period. Data reflect surveys of seven stands of
lodgepole pine in southern British Columbia between 2000 and 2005.

31

DISCUSSION
The increases in populations of species of secondary bark beetles one to two years
prior to the eruption of populations of mountain pine beetle in four of the five stands where a
population phase transition took place, in conjunction with the positive temporal correlations
between the number of trees colonized by secondary bark beetles early in the season and
montain pine beetle later in each season, suggest two mechanisms by which mountain pine
beetle may erupt. First, the increase in mountain pine beetle populations may be independent
of the numbers of trees colonized by secondary bark beetles. The increase may simply reflect
a delayed response to abiotic stresses that create increasingly favourable conditions for
reproduction for all insects. Berryman (1976), for example, suggested that a rapid decline in
stand resistance may trigger outbreaks of mountain pine beetle. Drought is one of many
stress events that can promote scolytid reproduction by increasing the susceptibility, and
perhaps nutritional quality, of host trees (Hopping and Mathers 1945, Rudinsky 1962,
Berryman 1972, Mattson and Haack 1987, Allen and Breshears 1998, Kelsey and Joseph
2001, Berg et al. 2006. RalTa et al. 2008). The increase in the number of trees colonized by
secondary bark beetles in this study is believed to be associated with an extended period of
low spring precipitation (Carroll et al. 2006). However, as in the southern pine beetle system
(Turchin 1991), there is conflicting evidence for drought as the solitary trigger of outbreaks of
mountain pine beetle, as populations have erupted in periods of below normal, normal, and
abundant precipitation (Blackman 1931, Beal 1943).
Second, the strong temporal relationships between secondary bark beetles and

32

mountain pine beetle, particularly the most populous bole-infesting species in this study, O.
latidens and P. mexicanus, lends support to a theory of facilitation in which a densitydependent facilitative relationship occurs exclusively at endemic levels of the population.
Amman and Schmitz (1988), for example, proposed that scattered populations of mountain
pine beetle must build up before an outbreak occurs. Species such as D. murrayanae, O.
latidens, P. mexicanus, Hyhirgops spp., and /. pini emerge and establish in hosts prior to
flight of mountain pine beetle. In turn, endemic populations of mountain pine beetle
emerging in late July, August, and early September have the opportunity to either seek out or
reject trees inhabited by these insects. A study by Moeck et al. (1981) suggested mountain
pine beetle does not reject trees that contain other species, such as D. valens (LeConte), D.
brevicomis, H. suhcostulatus (Mannerheim), O. latidens, and Pityophthorits serratus
(Swaine). Small-scale population buildups of species such as P. mexicanus, O. latidens, I.
pini, Hyhirgops spp., and D. murrayanae may enable the accumulation of endemic mountain
pine beetle by providing access to otherwise unsuitable hosts (see Smith et al. 2011). Once
populations of mountain pine beetle gain sufficient numbers to initiate mass attack of a
healthy host tree (i.e., approximately 300 to 500 beetles per hectare), they may no longer be
dependent on other bark beetles or weakened hosts, and begin to shift their colonization
behaviour accordingly (Carroll et al. 2006).
Strong interspecific competition could preclude facilitative relationships in this
system, however. For example, secondary species frequently outcompete more aggressive
primary bark beetles by rapid larval development (Rankin and Borden 1991). In weakened

33

and dying trees, secondary bark beetles are better suited to exploit available resources than
their primary bark beetle counterparts such as D. ponderosae or D. rufipennis (McCambridge
and Knight 1972, Poland and Borden 1998). While studies have provided compelling
evidence to suggest mountain pine beetle at epidemic levels would be better off minimizing
interspecific competition with secondary bark beetles such as /. pini (Rankin and Borden
1991, Safranyik et al. 1999), very little has been observed regarding endemic levels of the
insect. Endemic populations of mountain pine beetle are found in surprisingly low numbers,
estimated at less than forty beetles per hectare (Carroll et al. 2006). At such low numbers,
their density in a host would not reach levels reflective of outbreaks tested by Rankin and
Borden (1991) (i.e., 50 beetles/m2). Furthermore, Safranyik et al. (1999) suggested that high
levels of/, pini may, in some instances, enhance survival of mountain pine beetle by
accelerating the death of tissues in the upper part of the host.
Delayed density-dependent effects of predators and parisitoids, not uncommon in
defoliating systems (Myers 1988, Roland and Taylor 1997, Rothman and Roland 1998, Roland
2005, Dwyer ct al. 2004, Cooke and Lorenzetti 2006), that become diluted by increasing
populations of secondary bark beetles, could also facilitate increasing survivorship of
endemic populations of mountain pine beetle. Aggregation of the secondary bark beetle, /.
pini, for example, may dilute the effect of predation by generalist clerids and other bark beetle
predators (Aukema and Raffa 2004). Amman (1984) found that predation by clerids was
significantly higher in endemic populations of mountain pine beetle than in epidemic or postepidemic populations. While the largest contributors to brood mortality of mountain pine

34

beetle were cold over-wintering temperatures and desiccation, these factors did not differ
between infestation levels (Amman 1984), possibly placing greater significance on predation
as a mortality factor governing endemic populations. Although Boone et al. (2008) found
heterospecifics add to competition and predator load, the densities of infestation in their
experiments exceed levels found in endemic populations of mountain pine beetle (Carroll et
al. 2006).
Although Amman and Schmitz (1988) have outlined several predisposing factors for
an outbreak by mountain pine beetle, assessing the risk of outbreaks using hazard rating
systems that include these factors has not met with much success (Katovich and Lavigne
1985, Bentz et al. 1993, Nelson et al. 2008). Bentz et al. (1993) partially attributed the
inadequacy of hazard rating systems to a lack of knowledge concerning the endemic phase of
the insect, and Logan et al. (1998) suggested that spatial inference is necessary. Although
lacking evidence, for example, Amman and Schmitz (1988) believed there may be a close
relationship between secondary bark beetles and endemic levels of mountain pine beetle.
Likewise, Hamel and McGregor (1976) and Gohcen and Cobb (1980) suggested potential
associations between low level populations of mountain pine beetle and secondary bark
beetles, and Furniss and Carolin (1977) noted an association between endemic levels of
mountain pine beetle and D. brevicomis. The annual emergence patterns of secondaries and
mountain pine beetle in this study indicate that the latter were the last to enter host trees in a
season and would have had the opportunity to reject trees containing secondary species. The
next chapter examines spatial interactions between endemic levels of mountain pine beetle

35

and secondary bark beetles. Greater understanding of such interactions could provide new
management tools, such as the use of beetle monitoring to identify endemic populations of
mountain pine beetle on the verge of a population phase transition.

ACKNOWLEDGEMENTS
I thank B. S. Lindgren, F. R. McKee, and L. M. Poirier of the Forest Insect Research Group,
University of Northern British Columbia, for insight on this project. Funding for the project
was provided by the Natural Sciences and Engineering Research Council of Canada as well
as Genome BC. Data collection for this project was funded by the Government of Canada
through the Mountain Pine Beetle Initiative, a six-year, $40 million Program administered by
Natural Resources Canada, Canadian Forest Service.

36

LITERATURE CITED
Akaike, H. 1973. A new look at the statistical model identification. IEEE Transactions on
Automatic Control, 19: 716-723.
Allen, C. D., and D. D. Breshears. 1998. Drought induced shift of a forest-woodland ecotone:
Rapid landscape response to climate variation. Proceedings of the National Academy
of Sciences, 95: 14839-14842.
Amman, G. D. 1984. Mountain pine beetle (Coleoptera: Scolytidae) mortality in three types
of infestations. Environmental Entomology, 13: 184-191.
Amman, G. D., and R. F. Schmitz. 1988. Mountain pine beetle - lodgepole pine interactions
and strategies for reducing tree losses. Ambio, 17: 62-68.
Aukema, B. H., A. L. Carroll, J. Zhu, K. F. Raffa, T. A. Sickley, and S. W. Taylor. 2006
Landscape level analysis of mountain pine beetle in British Columbia, Canada:
Spatiotemporal development and spatial synchrony within the present outbreak.
Ecography 29: 427-441.
Aukema, B. H., and K. F. Raffa. 2004. Does aggregation benefit bark beetles by diluting
predation? Links between a group-colonization strategy and the absence of multiple
predator effects. Ecological Entomology, 29: 129-138.
Avery, T. E., and H. E. Burkhart. 2002. Forest measurements. 5th ed. McGraw-Hill, NY,
USA.
Baltcnsweilcr, W., and A. Fischlin. 1988. The larch budmoth in the Alps, pp. 331-351. In A.
A. Berryman (ed.), Dynamics of Forest Insect Populations: Patterns, Causes,
Implications. Plenum, New York.
Beal, J. A. 1943. Relation between tree growth and outbreaks of the Black Hills beetle.
Journal of Forestry, 41: 359-366.
Bentz, B. J., J. A. Logan, and G. D. Amman. 1991. Temperature-dependent development of
the mountain pine beetle (Coleoptera: Scolytidae) and simulation of its phenology.
The Canadian Entomologist, 123: 1083-1094.
Bentz, B. J., J. A. Logan, and G. D. Amman. 1993. A critical assessment of risk classification
systems for the mountain pine beetle. Forest Ecology and Management, 61: 349-366.

37

Berg, E. E., J. D. Henry, C. L. Fastie, A. D. De Voider, and S. M. Matsuoka. 2006. Spruce
beetle outbreaks on the Kenai Peninsula, Alaska, and Kluane National Park and
Reserve, Yukon Territory: relationship to summer temperatures and regional
differences in disturbance regimes. Forest Ecology and Management, 227: 219-232.
Berryman, A. A. 1972. Resistance of conifers to invasion by bark beetle-fungus associations.
Bioscience, 22: 598-602.
Berryman, A. A. 1976. Theoretical explanation of mountain pine beetle dynamics in
lodgepole pine forests. Environmental Entomology, 5: 1225-1233.
Blackman, M. W. 1931. The Black Hills beetle {Dendroctonus ponderosae Hopkins). N.Y
State College of Forestry, Syracuse, N.Y. Technical Publication No. 36.
Blais, J. R. 1965. Spruce budworm outbreaks in the past three centuries in the Laurentide
Park, Quebec. Forest Science, 11: 130-138.
Bright, D. E., Jr. 1976. The bark beetles of Canada and Alaska. Coleoptera: Scolytidae.
Canada Department of Agriculture, Biosystematic Research Institute Research
Branch, Ottawa, Ontario, Publication 1576. 241p.
Carroll, A. L., B. H. Aukcma, K. F. Raffa, G. D. Smith, and B. S. Lindgren. 2006. Mountain
pine beetle outbreak development: the endemic - incipient transition. Mountain Pine
Beetle Initiative, Project 1.03 Working Paper, Natural Resources Canada, Canadian
Forest Service. 21 p.
Carroll, A. L., S. W. Taylor, J. Regniere, and L. Safranyik. 2004. Effects of climate change on
range expansion by the mountain pine beetle in British Columbia, pp. 223-232. In
T. L. Shore, J. E. Brooks, and J. E. Stone (eds.), Mountain pine beetle symposium:
challenges and solutions. Natural Resources Canada, Canadian Forest Service, Pacific
Forestry Centre, Information Report BC-X-399, Victoria, British Columbia, Canada.
Cerezke, H. F. 1995. Egg gallery, brood production, and adult characteristics of mountain
pine beetle, Dendroctonus ponderosae Hopkins (Coleoptera: Scolytidae), in three pine
hosts. The Canadian Entomologist, 127: 955-965.
Cole, W. E., and G. D. Amman. 1969. Mountain pine beetle infestations in relation to
lodgepole pine diameters. Research Note INT-95. United States Department of
Agriculture, Forest Service, Intermountani Forest and Range Experiment Station,
Ogdcn, UT.

38

Cooke, B. J., and F. Lorenzetti. 2006. The dynamics of forest tent caterpillar outbreaks in
Quebec, Canada. Forest Ecology and Management, 226: 110-121.
Dwyer, G., J. Dushoff, and S. H. Yee. 2004. The combined effects of pathogens and predators
on insect outbreaks. Nature, 430: 341-345.
Furniss, M. M., and S. J. Kegley. 2008. Biology of Dendroctonus murrayanae (Coleoptera:
Curculionidae: Scolytinae) in Idaho and Montana and Comparative Taxonomic Notes.
Annals of the Entomological Society of America, 101: 1010-1016.
Furniss, M. M., and J. A. Schenk. 1969. Sustained natural infestation by the mountain pine
beetle in seven new Pinus and Picea hosts. Journal of Economic Entomology, 62:
518-519.
Furniss, R. L., and V. M. Carolin. 1977. Western Forest Insects. United States Forest Service
Miscellaneous Publication 1339. United States Department of Agriculture, Forest
Service, Washington, D. C.
Goheen, D. J., and F. W. Cobb Jr. 1980. Infestation of ceratocystis wageneri-infected
ponderosa pines by bark beetles (Coleoptera: Scolytidae) in the central Sierra
Nevada. The Canadian Entomologist, 112: 725-730.
Grove, S. J. 2002. Saproxylic insect ecology and the sustainable management of forests.
Annual Review of Ecology and Systematics, 33: 1-23.
Hamel, D. R.. and M. D. McGregor. 1976. Biological notes on the emergence of mountain
pine beetle and associates from lodgepole pine, Gallatin ranger district, Gallatin
National Forest, Montana 1975. Report 76-7. United States Department of Agriculture,
Forest Service, Northern Region, Missoula, MT.
Hopping, G. R., and W. G. Mathers. 1945. Observations on outbreaks and control of the
mountain pine beetle in the lodgepole pine stands of Western Canada. The Forestry
Chronicle, 21:98-108.
Johnson, D. M., A. M. Liebhold, O. N. Bj0rnstad, and M. L. McManus. 2005. Circumpolar
variation in periodicity and synchrony among gypsy moth populations. Journal of
Animal Ecology, 74: 882- 892.
Katovich, S. A., and R. J. Lavigne. 1986. The applicability of available hazard rating systems
for mountain pine beetle in lodgepole pine stands of southeastern Wyoming. Canadian
Journal of Forest Research, 16: 222-225.

39

Kelsey, R. G., and G. Joseph. 2001. Attraction of Scolytus unispinosus bark beetles to ethanol
in water-stressed Douglas-fir branches. Forest Ecology and Management, 144:
229-238.
Logan, J. A., P. White, B. A. Bentz, and J. A. Powell. 1998. Model analysis of spatial patterns
in mountain pine beetle outbreaks. Theoretical Population Biology, 53: 236-255.
Mattson, W. J., and R. A. Haack. 1987. The role of drought in outbreaks of plant-eating
insects. BioScience, 37: 110-118.
May, R. M. 1974. Biological populations with nonoverlapping generations: stable points,
stable cycles, and chaos. Science, 186: 645-647.
McCambridge, W. R, and F. B. Knight. 1972. Factors affecting spruce beetles during a
small outbreak. Ecology, 53: 830-839.
Miller, D. R., and J. H. Borden. 1985. Life history and biology of Ips latidens (LeConte)
(Coleoptera: Scolytidae). The Canadian Entomologist, 117: 859-871.
Moeck, H., D. L. Wood, and K. Q. Lindahl Jr. 1981. Host selection behaviour of bark beetles
(coleoptera: scolytidae) attacking Pinus ponderosa, with special emphasis on the
western pine beetle, Dendroctonus brevicomis. Journal of Chemical Ecology, 7:
49-83.
Myers, J. H. 1988. Can a general hypothesis explain population cycles of forest Lepidoptera?
Advances in Ecological Research, 18: 179-242.
Nelson, W. A., A. Potapov, M. A. Lewis, A. E. Hundsdorfcr, and F. He. 2008. Balancing
ecological complexity in predictive models: a reassessment of risk models in the
mountain pine beetle system. Journal of Applied Ecology, 45: 248-257.
0kland, B., and O. N. Bj0rnstad. 2006. A resource-depletion model of forest insect
outbreaks. Ecology, 87: 283^290.
Poland, T. M., and J. H. Borden. 1998. Competitive exclusion of Dendroctonus rufipennis
induced by pheromones of Ips tridens and Dryocoetes ajfaber (Coleoptera:
Scolytidae). Journal of Economic Entomology, 91: 1150-1161.
Raffa, K. F , B. H. Aukema, B. J. Bentz, A. L. Carroll, J. A. Hicke, M. G. Turner, and W. H.
Romme. 2008. Cross-scale drivers of natural disturbances prone to anthropogenic
amplification: the dynamics of bark beetle eruptions. Bioscience, 58: 501-517.
doi:10.164l/B580607.
40

Raffa, K. F., and A. A. Berryman. 1983. The role of host plant resistance in the colonization
behavior and ecology of bark beetles (Coleoptera: Scolytidae). Ecological
Monographs, 53: 27-49.
Rankin, L. J., and J. H. Borden. 1991. Competitive interactions between the mountain pine
beetle and the pine engraver in lodgepole pine. Canadian Journal of Forest Research,
21: 1029-1036.
Rasmussen, L. A. 1974. Flight attack and behavior of mountain pine beetles in lodgepole
pine of northern Utah and southern Idaho. Research Note INT-180. United States
Department of Agriculture, Forest Service, Intermountain Research Station, Ogden,
UT.
Reid, R. W. 1963. Biology of the mountain pine beetle, Dendroctonus monticolae Hopkins, in
the East Kootenay region of British Columbia III. Interactions between the beetle and
its host, with emphasis on brood mortality and survival. Canadian Entomologist, 95:
225-238.
Regniere, J., and B. J. Bentz. 2007. Modeling cold tolerance in the mountain pine beetle,
Dendroctonus ponderosae. Journal of Insect Physiology, 53: 559-572.
Roland, J. 2005. Are the "seeds" of spatial variation in cyclic dynamics apparent in
spatially-replicated short lime-series? An example from the forest tent caterpillar.
Annales Zoologic Fennici, 42: 397-407.
Roland, J., and R D. Taylor. 1997. Insect parasitoid species respond to forest structure at
different spatial scales. Nature, 386: 710-713.
Rotlmian, L. D., and J. Roland. 1998. Forest fragmentation and colony performance of forest
tent caterpillar. Ecography, 21: 383-391.
Rudinsky, J. A. 1962. Ecology of Scolytidae. Annual Review of Entomology, 7: 327-348.
Safranyik, L., and A. L. Carroll. 2006. The biology and epidemiology of the mountain pine
beetle in lodgepole pine forests, pp. 3-66. In L. Safranyik and B. Wilson (eds.), The
mountain pine beetle: a synthesis of its biology, management and impacts on
lodgepole pine. Pacific Forestry Centre, Canadian Forest Service, Natural Resources
Canada, Victoria, British Columbia.

41

Safranyik, L., A. L. Carroll, J. Regniere, D. W. Langor, W. G. Riel, T. L. Shore, B. Peter, B. J.
Cooke, V. G. Nealis, and S. W. Taylor. 2010. Potential for range expansion of
mountain pine beetle into the boreal forest of North America . The Canadian
Entomologist, 142: 415-442.
Safranyik, L., and D. A. Linton. 1998. Mortality of mountain pine beetle larvae,
Dendroctonus ponderosae (Coleoptera: Scolytidae) in logs of lodgepole pine (Pinus
contorta var. latifolia) at constant low temperatures. Journal of the Entomological
Society of British Columbia, 95: 81-87.
Safranyik, L., D. A. Linton, and T. L. Shore. 2000. Temporal and vertical distribution of bark
beetles (Coleoptera: Scolytidae) captured in barrier traps at baited and unbaited
lodgepole pines the year following attack by mountain pine beetle. The Canadian
Entomologist, 132: 799-810.
Safranyik, L., T. L. Shore, A. L. Carroll, and D. A. Linton. 2004. Bark beetle (Coleoptera:
Scolytidae) diversity in spaced and unmanaged mature lodgepole pine (Pinaceae) in
southeastern British Columbia. Forest Ecology and Management, 200: 23-38.
Safranyik, L., T. L. Shore, D. A. Linton and L. Rankin. 1999. Effects of induced competitive
interactions with secondary bark beetle species on the establishment and survival of
mountain pine beetle broods in lodgepole pine. Information Report BC-X-384, Pacific
Forestry Centre, Canadian Forest Service, Natural Resources Canada, Victoria,
British Columbia. 40p.
Safranyik, L., D. M. Shrimpton, and H. S. Whitney. 1974. Management of lodgepole pine to
reduce losses from the mountain pine beetle. Forestry Technical Report 1, Pacific
Forestry Centre, Canadian Forest Service, Natural Resources Canada, Victoria,
British Columbia. 29p.
Sambaraju, K. R., A. L. Carroll, J. Zhu, K. Stahl, R. D. Moore, and B. H. Aukcma. 2011.
Climate change could alter the distribution of mountain pine outbreaks in western
Canada. Ecography, In review.
Schenk, J. A., and D. M. Benjamin. 1969. Notes on the biology of Ips pini in central
Wisconsin jack pine forests. Annals of the Entomological Society of America, 62:
480-485.
Smith, G. D., A. L. Carroll, and B. S. Lindgren. 2009. Life history of a secondary bark
beetle, Pseudips mexicamis (Coleoptera: Curculionidae: Scolytinac), in lodgepole
pine in British Columbia. The Canadian Entomologist, 141: 56-69.

42

Smith, G. D., A. L. Carroll, and B. S. Lindgren. 2011. Facilitation in bark beetles: endemic
mountain pine beetle gets a helping hand. Agricultural and Forest Entomology, 13:
37-43 .
Smith, R. H., J. P. Cramer, and E. J. Carpender. 1981. New record of introduced hosts for the
mountain pine beetle in California. Research Note PSW-354. United States
Department of Agriculture, Forest Service, Pacific Southwest Forest & Range
Experiment Station, Berkeley, CA.
Turchin, P., P. L. Lorio Jr., A. D. Taylor, and R. F. Billings. 1991. Why do populations of
southern pine beetles (Coleoptera: Scolytidae) fluctuate? Environmental Entomology,
20: 401-409.
Westfall, J., and T. Ebata. 2009. 2008 summary of forest health conditions in British
Columbia. Ministry of Forests and Range, Forest Practices Branch. British Columbia
Forest Service, Victoria, British Columbia, Canada.
Williams, D. W., and A. M. Liebhold. 1995. Detection of delayed density dependence: effects
of autocorrelation in an exogenous factor. Ecology, 76: 1005-1008.
Wood, S. L. 1982a. The role of pheromones, kairomones, and allomones in the host selection
and colonization behavior of bark beetles. Annual Review of Entomology, 27:
411-446.
Wood, S. L. 1982b. Editor. The bark and ambrosia beetles of North and Central America
(Coleoptera: Scolytidae), a taxonomic monograph. Brigham Young University, Provo,
Utah. 1359p.
Wulder, M. A., C. C. Dymond, J. C. White, D. G. Leckie, and A. L. Carroll. 2006. Surveying
mountain pine beetle damage of forests: A review of remote sensing opportunities.
Forest Ecology and Management 221: 27-41.

43

CHAPTER 3
Spatial associations of mountain pine beetle, Dendroctonus ponderosae, with secondary bark
beetles in the endemic to incipient-epidemic phase transition
ABSTRACT
The mountain pine beetle, Dendroctonus ponderosae Hopkins, is native to western
North America and attacks most species of pine in its range. Its population dynamics are
characterized by four phases: endemic, incipient-epidemic, epidemic, and post-epidemic.
Beetles typically subsist at endemic levels for many years between outbreaks, reproducing in
the tissues of weakened or dying trees. Very little attention has been paid to populations at
endemic stages, because they do not kill large numbers of healthy trees. This study explored
the stand-level spatial interactions of endemic beetles with other bark beetles frequently
found in weakened pine hosts. Endemic and incipient-epidemic levels of mountain pine
beetle were often positively spatially associated with secondary bark beetles such as Pseudips
mexicanus (Hopkins), Orthotomicus latidens (LeConte), Ips pini (Say), Hylurgops porosus
(LeConte), and H. rugipennis (Mannerheim). As populations grew, mountain pine beetle
shifted from attacking injured and previously colonized hosts to more vigorous hosts in a
clustered pattern. The positive spatial associations may indicate a facilitative relationship
between endemic mountain pine beetle and other phloem-infesting bark beetles, and provide
insight into mechanisms potentially facilitating the transition of the organism from the
endemic to incipient-epidemic phase.
Key words: phase transition; facilitation; competition; niche partitioning

44

INTRODUCTION
The mountain pine beetle, Dendroctonus ponderosae Hopkins, is a bark beetle that
exhibits a broad geographic range extending from the upper limits of Mexico northward to
northwestern British Columbia, and from the Pacific coast eastward to South Dakota in the
United States, and eastern Alberta in Canada (Safranyik and Carroll 2006). A generalist,
mountain pine beetle is capable of colonizing nearly every species of native and introduced
pine within its range, although lodgepole pine, Pinus contorta (Douglas ex Louden), is
considered its principal host (Furniss and Schenk 1969, Smith et al. 1981, Cerezke 1995,
Carroll et al. 2004, Safranyik and Carroll 2006). The population dynamics of this insect
consists of four phases; endemic, incipient-epidemic, epidemic, and post-epidemic or
collapse (Safranyik and Carroll 2006).
Mountain pine beetles are typically found at endemic levels, where populations
experience significant mortality due to cold winter temperatures and larval desiccation
(Amman 1984). Favourable conditions for beetle reproduction include successive warm
summers, mild winters, and stress events such as drought (Reid 1963, Safranyik et al. 1974,
Thomson and Shrimpton 1983). When such conditions coincide with an adequate number of
mature hosts, the beetle may enter the incipient-epidemic phase (Carroll et al. 2006). From
the incipient-epidemic phase, beetle populations may progress to outbreak status, where mass
attacks coordinated by aggregation pheromones enable the beetles to overwhelm the defenses
of healthy, large diameter trees (Wood 1982, Raffa and Berryman 1983). Mountain pine
beetles are aided in overcoming the defenses of host trees by innoculating host tissue with

45

mutualistic phytopathogenic fungi (Francke-Grossman 1967, Berryman 1972) transported in
specialized cuticular structures called mycangia (Batra 1963, Paine et al. 1997). Fungal
growth not only disrupts the defensive capacity of host trees, but may also provide nutritional
benefits for phloem-feeding larvae (Bleiker and Six 2007).
At outbreak levels, the insects can kill trees over vast regions, exhibiting biome-level
consequences (Aukema et al. 2006, Raffa et al. 2008, Kurz et al. 2008). Outbreaks typically
collapse when the beetles exhaust the available large-diameter host resources, and/or cold
winter temperatures induce sufficient brood mortality to reduce reproduction below the
replacement numbers required to sustain an epidemic (Reid 1963, Cole and Amman 1969,
Safranyik et al. 1974, Amman 1984). Mountain pine beetle populations thereby return to an
endemic level where they will remain, often for decades, until favourable conditions arise
again.
Our understanding of epidemic populations of mountain pine beetle is considerable,
yet little is known about endemic populations or the transition of populations from endemic
to incipient-epidemic levels. Even quantification of such populations remains challenging. It
was previously thought, for example, that endemic beetles may be found infesting less than
one tree in 40.5 hectares of forest (Amman 1984). However, Amman and Schmitz (1988)
later referred to endemic populations as those comprising less than one mass-attacked tree
per ten hectares. More recently, Carroll et al. (2006) estimated the number of endemic
beetles to be less than forty individuals per hectare, and incapable of mass-attacking even a
single tree.

46

If endemic beetles are unable to overcome the defences of healthy trees (Raffa and
Berryman 1983), they may rely on the presence of other bark beetles as well as other insects
and diseases in marginal hosts to facilitate reproduction until the establishment of sufficiently
viable populations enables them to overcome healthy trees (Carroll et al. 2006). This
hypothesis is not without anecdotal evidence in the literature. For example, endemic levels of
mountain pine beetle may utilise a broad spectrum of weakened hosts, and association with
other bark beetles has been suggested by DeLeon (1934), Hamel and McGregor (1976), and
Goheen and Cobb (1980). Moreover, previous studies have implied positive associations
between mountain pine beetle and other host-stressing agents such as the root disease
Armillaria mellea (Hinds et al. 1984, Lessard et al. 1985, Tkacz and Schmitz 1986), and
dwarf mistletoe Arceuthobium spp. (McCambridge et al. 1982, Rasmussen 1987)
In this study I examined the stand-level spatial associations between endemic-level
mountain pine beetle and other bark beetles including Pseudips mexicanus (Hopkins),
Orthotomicus latidens (LeConte), Ips pini (Say), Hyiurgops porosus (LeConte), H.
rugipennis (Mannerheim) and D. murrayanae (Hopkins) over five years in seven lodgepolc
pine stands near Mcrritt and Kelowna, British Columbia. In five of these stands, mountain
pine beetle transitioned from endemic to outbreak levels over the course of the study. I tested
whether mountain pine beetle exhibits spatial dependencies with associated beetles through
time, which would suggest that a suite of bole-infesting bark beetles may be key contributors
to the development of incipient-epidemic populations of mountain pine beetle.

47

METHODS
Refer to Chapter 2, Methods: Study Sites, for source of data. Maps for the locations
of all trees inhabited by mountain pine beetle and secondary bark beetles were created using
the 'spatstat' package v. 1.13-3 in R v.2.6.2 (Ihaka and Gentleman 1996, Baddeley and Turner
2005, R Development Core Team 2008). Colonized trees were categorized by their
respective year and species. Trees that were colonized by mountain pine beetle were further
categorized as: resisted attack (where the beetles were pitched out by the tree), strip attack
(where only a portion or strip of the bole was attacked), and mass attack (where the majority
of the bole was attacked). Logistic regression was used to test Rasmussen's hypothesis (1974)
that early emerging mountain pine beetle are more likely to attack trees with prior injury by
analysing attacked/unattacked trees as a function of injury status.
Spatial point process models were used to evaluate the spatial relationships between
trees colonized by mountain pine beetle and those colonized by other bark beetles. The other
bark beetles were chosen if they exhibited temporal relationships with mountain pine beetle
(Chapter 2). Analyses were conducted when there were two or more instances of
colonization by both mountain pine beetle and the secondary species in the stand in that year.
New techniques in spatial point process modeling are powerful for discerning
potential relationships between species or other biotic and abiotic factors that may otherwise
go undetected (Stoyan and Penttinen 2000). Spatial point process models were used to
examine the influence of previous years of colonization by other bark beetles on the varying
types of mountain pine beetle attack. Furthermore, point process models also examined the

48

effects of these insects colonizing trees the same year as mountain pine beetle, since they
flew and were recorded prior to peak mountain pine beetle flight in July and August. Finally,
the number of trees mass-attacked by mountain pine beetle were examined with respect to
strip attacks in years t and t-\.
The response variable for each model, a spatially-explicit estimated density of trees
colonized by mountain pine beetle (A), was measured as the number of trees bearing attacks
by mountain pine beetle (either resisted attack, strip attack, mass attack, or a combination)
per square meter of stand area. Covariates (here, the location of attacks by other bark beetles)
were converted from point locations to density surfaces prior to fitting. This process
incorporates a Gaussian kernel density smoother as a representation of the point process
defined within the boundary of each stand (Cressie 1991, Baddeley and Turner 2000). A
periodic border correction was tested for a subset of models, but not utilised, as results proved
robust to methods with a default edge border correction.
Parameters in these spatial point process regression models were estimated using
maximum pseudolikclihood methods. Significance of individual variables was judged by
statistical comparison to a homogenous model, i.e., one estimating only an intercept or a
constant intensity of resisted, strip, or mass attack by mountain pine beetle across the site, by
examining the change in deviance relative to a % 2 reference distribution. Where I sought to
examine additive effects of multiple variables, a comparison of nested models was performed
by examining the change in deviance relative to a x, 2 reference distribution. Models were
compared using Akaike's Information Criterion (AIC), and models with the lowest AIC

49

values were judged to fit best (Akaike 1973).
Spatial analyses to determine the extent of clustering of strip and mass attack by
mountain pine beetle were performed using Ripley's K functions. Simulation envelopes were
used to construct a 95% confidence interval (n - 999 simulations). Trees were judged to be
"clustered" if falling above the upper limits of this interval about the empirical function, but
spatially "inhibited" if below the interval (Ripley 1981).
RESULTS
The populations of mountain pine beetle in five of stands (denoted A, B, C, F and E)
underwent a transition from endemic to incipient-epidemic levels over the course of the
study, as the number of trees strip- and mass-attacked by mountain pine beetle increased each
year in these four stands (Tables 3.1 and 3.2). In stands in which populations of mountain
pine beetle did not erupt (D and G), the number of trees strip- and mass-attacked by mountain
pine beetle reached a small peak in 2004 and dropped substantially the following year (Tables
3.1 and 3.2). The number of trees strip- and mass-attacked by mountain pine beetle in stand
E increased from 2002 to 2003 (Tables 3.1 and 3.2). Due to the large increase in populations
between these years (likely due to immigration and not endogenous stand dynamics), this
stand was dropped from censusing after 2003.
Many of the trees strip-attacked by mountain pine beetle prior to the population phase
transition showed evidence of colonization by other species of bark beetle and/or some form
of injury such as a broken or forked top, a scarred trunk, a thin crown, an infection of dwarf
mistletoe (Arceuthobium spp.), an attack by mountain pine beetle greater than ten years prior,

50

or some other form of injury (Table 3.1; see Carroll et al. (2006) for a more detailed list of
injuries). For the stands in which a phase transition from endemic to incipient-epidemic
levels was occurring (i.e., stands A, B, C, and F), the number of trees strip-attacked by
mountain pine beetle with injuries and/or secondary colonization began to decline in the final
year of the study. During the transition from endemic to incipient-epidemic levels, many
mass-attacked trees were also previously colonized by secondaries and/or possessed some
form of injury (Table 3.2). Once populations of mountain pine beetle were established at the
incipient-epidemic level however, the insects rarely mass-attacked trees that had been
previously colonized by other species of bark beetles.

51

Table 3.1: Number of lodgepole pine trees strip-attacked by mountain pine beetle in seven
stands in southern British Columbia, 1999-2005. Within the subset of these trees, the
numbers bearing injuries and/or any one or multiple colonizations of other bark beetles
(Dendroctonus murrayanae, Hylurgops spp., Ips pini, Orthotomicus latidens, Pseudips
mexicanus) are listed.
Stand
No. Trees
Year
1999
2000
2001
2002
2003
2004
2005
A
Strip attacks
4
9
17
25
37
32
70
Injured
4
9
17
22
18
14
9
Other bark beetles
0
3
10
18
11
10
10
B

Strip attacks
Injured
Other bark beetles

C

Strip attacks
Injured
Other bark beetles

D

3
2
0

11
8
3

27
24
16

37
30
19

52
36
9

2
2
1

5
4
4

5
5
3

4
3

21
5
1

65
6
1

14
9
10

17
7
12

3

Strip attacks
Injured
Other bark beetles
Strip attacks
Injured
Other bark beetles

13
9
0

16
13
5

61
21
5

Strip attacks
Injured
Other bark beetles

1
0
0

2
1
0

18
6
7

25
12
12

41
8
7

Strip attacks
Injured
Other bark beetles

-

4
3
2

12
3
8

21
7
18

2
0
1

-

52

Table 3.2: Number of lodgepole pine trees mass-attacked by mountain pine beetle in seven
stands in southern British Columbia, 1999-2005. Within the subset of these trees, the
numbers bearing injuries and/or any one or multiple colonizations of other bark beetles
(Dendroctonus murrayanae, Hylurgops spp., Ips pini, Orthotomicus latidens, Pseudips
mexicanus) are listed.
Stand
Year
No. Trees
1999
2002
2004
2005
2000
2001
2003
A
Mass attacks
0
1
6
99
296
6
65
Injured
0
5
33
28
1
4
33
4
Other bark beetles
0
1
4
11
0
3
B

C

D

E

F

G

Mass attacks
Injured
Other bark beetles

1
1
0

0
0
0

3
3
2

38
25
8

129
83
11

_

_

-

-

Mass attacks
Injured
Other bark beetles

_

2
2
2

2
2
2

2
1
1

23
6
4

45
6
1

205
14
0

Mass attacks
Injured
Other bark beetles

-

0
0
0

0
0
0

5
4
3

11
9
7

11
4
6

6
0
4

Mass attacks
Injured
Other bark beetles

_

5
3
2

36
18
9

322
68
9

-

_

-

0
0
0

-

-

Mass attacks
Injured
Other bark beetles

-

_

-

-

0
0
0

5
3
2

24
13
11

47
19
9

191
44
1

_

0
0
0

0
0
0

0
1

11
6
6

14
3
6

4
1
2

Mass attacks
Injured
Other bark beetles

-

-

-

Emergence by mountain pine beetle in this study prior to late-July flight periods was
rare; less than 2% of all trees colonized by mountain pine beetle were attacked between June
and mid-July. The likelihood that beetles attacked an injured host was higher with beetles
that flew earlier vs. later (71.4% vs. 34.2%, P(attack injured host) = exp 092157x /l+exp 092 157x,
where x = 1 if late and 0 otherwise; Z2576 = -4.57 for estimate of time coefficient; P <0.0001).
Spatial Analyses
Mountain pine beetle and other bark beetles in the bole-infesting assemblage were
found occupying the same, or nearby, host trees in all stands. Significant spatial associations
between trees bearing strip-attack by mountain pine beetle and secondary bark beetle
colonization were generally uniform throughout stands undergoing a population eruption and
are summarized using stand B as a representative case (Table 3.3). Due to the extensive
nature of this study, the remaining stands have been placed in Appendix C (see Appendix C:
Tables C.1-C.6). In the two stands where mountain pine beetle did not erupt to epidemic
population levels, other bark beetles such as P. mexicanus, and O. latidens were often found
in close proximity with mountain pine beetle, sometimes even sharing the same host tree
(Appendix C: Tables C.3, C.6). These spatial associations, however, were not as numerous as
those found in stands undergoing a population phase transition (Appendix C: Tables C.l, C.2,
C.4, and C.5).

54

Table 3.3: Association of trees colonized by other bark beetles on the locations of trees stripattacked by mountain pine beetle from 2000 to 2003 in a lodgepole pine stand of southern
British Columbia (Stand B). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. For example, the estimated density of strip attack by mountain
pine beetle in 2001 in locations where all secondaries colonized trees at a rate of 0.0005/m2
or 5 trees/ha is e x p ( 1146 + 38™*o°o°5) = 0.0001 or 1 tree/ha. Significant models are listed in order
of best fit for each year.
2
Insect
Intercept
P-value
Year
Slope
AIC

x

2000
2000

Est. SE
-10.05 0.30
-11.46 0.88

Strip attack
2001
All secondaries 2001
H. spp.
2001

-9.15 0.19
-11.14 1.01
-10.51 0.74

3870 1849
7177 3568

Strip attack
All secondaries
P. mexicanus
O. latidens
P. mexicanus
All secondaries

2002
2002
2002
2002
2001
2001

-8.84
-10.08
-9.98
-9.99
-10.22
-10.47

0.16
0.63
0.60
0.61
0.71
0.85

1506
2699
4530
5461
3201

Strip attack
H. spp.
H. spp.
P. mexicanus
I. pini
All secondaries
All secondaries
P. mexicanus
O. latidens

2003
2003
2002
2003
2003
2003
2002
2002
2003

-8.50
-11.46
-10.57
-11.17
-10.70
-10.95
-10.96
-10.51
-9.49

0.14
0.6 i
0.42
0.62
0.49
0.57
0.66
0.61
0.34

28745
17534
8839
32277
3865
2846
4585
6264

Strip attack
P. mexicanus

Est.

SE
0.03

245.11
242.58

4.40
4.29

0.04
0.04

550.22
547.73
547.84

696
1274
2196
2605
1579

5.21
5.15
4.63
4.48
4.11

0.02
0.02
0.03
0.03
0.04

729.95
726.62
726.68
727.20
727.34
727.71

5016
2869
1770
6007
772
685
1251
1709

44.11
38.89
34.13
32.69
32.63
21.63
17.35
13.88

<0.0001
<0.0001
<0.000l
<0.0001
<0.000l
<0.0001
<0.0001
0.0002

989.68
947.38
952.61
957.37
958.80
958.86
969.86
974.14
977.61

31786 16262

55

4.49

In all stands, the locations of trees strip-attacked by mountain pine beetle were
associated with all trees colonized by other bark beetles considered as a whole. Cohabitation
or host sharing between endemic levels of mountain pine beetle and all secondary bark
beetles is exemplified by stand A in 2002 (Fig. 3.1).
Often, however, the locations of trees strip-attacked by mountain pine beetle could be
predicted by knowing the locations of trees colonized by only one species of bark beetle,
rather than the entire complex. For example, the locations of strip-attacked trees were
associated most frequently with colonization by P. mexicanus (e.g., similar AIC values for
models with all secondaries vs. P. mexicanus; Table 3.3: 2000, 2002, 2003). However, trees
strip attacked by mountain pine beetle also appeared to be positively associated with
colonization by Hylurgops spp., primarily in the earliest years of study (Table 3.3: 2001;
Appendix C: Tables C.l, C.2, C.4 and C.5), and with the locations of O. latidens (Table 3.3:
2002 and 2003; Appendix C: Tables C.l, C.2, C.5 and C.6). Fewer significant spatial
associations between trees colonized by /. pint, and D. murrayanae and those strip-attacked
by mountain pine beetle were apparent, however.

56

x = other bark beetles 2002
o =D ponderosae 2002

%>

<

340m
0

15

*

>

X X

>
o
x

-,

770m
Figure 3.1: Locations of trees strip attacked by mountain pine beetle, and colonizations by
Dendroctonus murrayanae, Hylurgops spp., Ips pint, Orthotomicus latidens, and Pseudips
mexicanus in southern British Columbia, stand A, 2002. Colonizations by mountain pine
beetle and other bark beetles comprise approximately 0.12 and 1.3% of the 19,500 pine trees
in stand A respectively.

57

Although most associations between mountain pine beetle and other bark beetles were
positive, strip attacks by mountain pine beetle following the endemic to incipient-epidemic
transition in stand A were negatively associated with /. pini, P. mexicanus and Hylurgops
spp. (Appendix C: Table C.l) in the final years of study. In stand E, there was also a weak
negative association between P. mexicanus and trees that were strip-attacked by mountain
pine beetle in 2002 (Appendix C: Table C.4).
Trees that were strip-attacked in one year were frequently mass-attacked by mountain
pine beetle the following year, as there were significant positive associations between stripattacked and mass-attacked trees (Table 3.4). In stands where no population phase transition
of mountain pine beetle was apparent (stands D and G), there were no significant associations
between strip attack and mass attack by mountain pine beetle.
In 2001, when populations of mountain pine beetle were still at endemic or early
incipient-epidemic levels, trees that exhibited strip attacks, although closely associated with
other bark beetles, were scattered in a random pattern throughout the stand (empirical line of
Ripley's K function is within simulation envelope. Fig. 3.3). However, as populations of
mountain pine beetle began to transition from endemic to incipient-epidemic levels, trees that
were strip- or mass-attacked began to be found in clusters (visual representation in Fig. 3.2,
Ripley's K function above simulation envelope in 3.4, and see Appendix C: Figs. C.l-C.4).
Clustering, of strip- and mass-attacked trees as the outbreak progressed was very pronounced
in all stands except two (results not shown; stands D and G).

58

Table 3.4: Association of trees strip-attacked by mountain pine beetle on the location of mass
attacks from 2002 to 2005 in a lodgepole pine stand of southern British Columbia (Stand B).
The line in bold represents an intercept-only model; i.e., modeling a constant density of
insects across the stand. Subsequent lines reflect whether the location of each listed insect
and year provides inference on the location of the insect studied relative to this constant
density. A positive estimate for a slope reflects positive spatial association, while a negative
estimate reflects spatial inhibition at a between-tree scale. The response variable for each
equation is log(A), where A is the estimated density of trees colonized per square meter.
Significant models are listed in order of best fit for each year.
Insect

Year

x1

Mass attack
Strip attack
Strip attack

2002
2002
2001

Intercept
Est. SE
-8.81 0.16
-11.24 0.68
-10.74 0.63

Slope

Mass attack
Strip attack
Strip attack

2003
2003
2002

-7.59
-8.85
-9.88

0.09
0.20
0.36

Mass attack
Strip attack
Strip attack

2004
2003
2004

-6.17
-6.78
-6.57

0.04
0.09
0.07

2418 266
8407 1123

79.08
52.22

<0.0001
<0.0001

7620.12
7541.18
7568.04

Mass attack
Strip attack
Strip attack

2005
2005
2004

-4.93
-6.62
-5.23

0.02
0.07
0.04

949
6479

32
615

959.50
104.5

<0.0001
<0.0001

21915.37
20953.26
21808.26

Est.

13798 3385
15449 4494

4557 530
13067 1801

59

P-value

AIC

<0.0001
0.0002

747.60
729.12
735.92

<0.0001
<0.0001

2217.79
2147.39
2155.71

SE
20.35
13.55

71.95
63.62

x = strip attack 2002
o = mass attack 2003

xo °

x
X

%

0
X
0

«o 0

340m

0

0

i*

0

0

0

0
X

x
0

X

X

<

<

X

rfo
X

770m
Figure 3.2: Location of trees strip attacked by mountain pine beetle in 2002, and mass
attacked in 2003 in southern British Columbia, stand A. Strip and mass attacks comprise
approximately 0.12 and 0.33% of the 19,500 lodgepole pine trees in stand A respectively.

60

uppci

— — — — theoretical

•i\1
0

cJZ--""
1
20

1
40

_ —

1
60

-T80

Figure 3.3: Ripley's K function for trees strip-attacked by mountain pine beetle in 2001 for
stand A. Observed estimate is shown by the black solid line, the upper and lower limits of the
95% confidence interval are shown by the green and blue dashes respectively. The theoretical
estimate for a point process displaying complete spatial randomness is shown by the red
dashes. The focal distance (r) on the x-axis is represented in metres.

61

, . . # « *.

uppci

*— •— *

lowei

—. — — ihcoicli^al

_

20

_

J

-r -

-r-

40

60

Figure 3.4: Ripley's K function for trees strip-attacked by mountain pine beetle in 2004 for
stand A. Observed estimate is shown by the black solid line, the upper and lower limits of the
95% confidence interval are shown by the green and blue dashes respectively. The theoretical
estimate for a point process displaying complete spatial randomness is shown by the smooth
red dashes. The focal distance (r) on the x-axis is represented in metres.

62

DISCUSSION
Colonization of hosts by bark beetles, particularly endemic-level mountain pine
beetle, poses an ecological paradox (Light et al. 1983). At epidemic levels, mountain pine
beetle reproduce in the tissues of vigorous hosts; however, such hosts are unavailable to the
insects at endemic levels (Raffa and Berryman 1983). Therefore, endemic mountain pine
beetle appear to rely on hosts unable to mount defensive responses capable of displacing the
insects. Such hosts may include trees with root disease (e.g., Armillaria spp.), or dwarf
mistletoe (e.g., Arceuthobium spp.) (McCambridge et al. 1982, Hinds et al. 1984, Lessard et
al. 1985, Tkacz and Schmitz 1986, Rasmussen 1987). Although the level of root rot in our
stands was low, the level of dwarf mistletoe was quite high. Many trees colonized by
endemic mountain pine beetle possessed some form of injury or disease, with the most
common being a broken top or dwarf mistletoe.
Colonizing trees with injury or disease poses a challenge, however. Weakened trees
are often infested by secondary bark beetles (Amman and Schmitz 1988), such that endemic
populations of mountain pine beetle and secondary bark beetles frequently inhabit the same
types of hosts (Bartos and Schmitz 1998). In this study, not only did mountain pine beetle
and secondary bark beetles inhabit the same types of hosts, but they frequently inhabited
hosts together (Fig. 3.1). Selecting hosts that offer the best opportunity for survival may
involve a degree of compromise where the probability of survival in a poorly defended host is
higher, but the mortality cost due to interspecific competition is potentially higher as well.
The presence of such a diverse and abundant assemblage of bark beetle species in all

63

stands may be an indicator of relaxed levels of competition, however (Ratchke 1976).
Competition can be reduced through strategies such as niche partitioning, for example (Byers
1989, Raffa 1991, Schlyter and Anderbrant 1993, Amezaga and Rodriguez 1998, Ayres et al.
2001). Amman and Schmitz (1988) suggest that the lower 30 to 60 cm of the bole may be
freely available to mountain pine beetle when associated with other bole-infesting bark
beetles. Moreover, Ayres et al. (2001) suggest that high numbers of interspecific associations
may benefit the rarest species, such as endemic levels of mountain pine beetle in the present
example.
This study provides evidence that the effects of competition as a mortality factor may
be most pronounced as populations of mountain pine beetle transition to epidemic levels
versus remaining at the endemic level. For example, there were only a few significantly
positive spatial associations found between mountain pine beetle and /. pini, and the two
insects colonized different hosts once mountain pine beetle entered the incipient-epidemic
phase (Appendix C: Table C.l). This is consistent with observations that mountain pine
beetles at epidemic levels may be outcompcted by /. pini (Bergvinson and Borden 1991,
Rankin and Borden 1991, Safranyik et al. 1999, Boone et al. 2008).
In contrast, endemic populations of mountain pine beetle may benefit from close
associations with bole-infesting heterospecifics. The majority of trees co-colonized by
mountain pine beetle and other bark beetles contained P. mexicanus, Hylurgops spp., and/or
O. latidens species. Seasonal phenologies of these bark beetle species (Chapter 2; Schenk
and Benjamin 1969, Miller and Borden 1985, Safranyik et al. 2004, Furniss and Kcgley 2008,

64

Smith et al. 2009) indicate that mountain pine beetle colonized these trees following
colonization by heterospecifics. The benefits of colonizing trees with established
heterospecifics may include increased nutritional quality of the host (Hodges et al. 1968,
Ayres et al. 2000, Bleiker and Six 2007), favourable moisture regulation due to extensive
fungal innoculation (Reid 1963, Whitney 1971, Amman 1977), reduced likelihood of
predation (Abrams et al. 1998, Ayres et al 2001, Aukema and Raffa 2004), and/or decreased
probability of mortality due to exhaustion of host defenses (Christiansen et al. 1987, Carroll
et al. 2006). For example, Ips spp. may colonize diseased trees prior to mountain pine beetle,
further weakening the host and/or altering host physiology, potentially resulting in the
production of chemicals attractive to Dendroctonus species (Hodges et al. 1968, Goheen and
Cobb 1980). Moreover, Smith (2008) found that endemic level populations of mountain pine
beetles reared with P. mexicanus in naturally infested host tissues developed more quickly,
produced more offspring, and were not significantly different in size, than those reared on
their own. Carroll et al. (2006) also found that phloem consumption by endemic mountain
pine beetle was positively influenced by phloem consumption of secondary bark beetle
species belonging to the bole-infesting assemblage.
In summary, these results are consistent with a model in which the colonization
dynamics of mountain pine beetle change as populations increase. Early emerging beetles
attack hosts with injury or attack from a previous year (Rasmussen 1974), particularly in
endemic populations where very few conspecifics are present (Reid et al. 1967).
Colonization of such trees, co-colonized by other bole-infesting bark beetles (Hamel and

65

McGregor 1976, Furniss and Carolin 1977, Goheen and Cobb 1980, Amman and Schmitz
1988, Carroll et al. 2006, Safranyik and Carroll 2006) may enable the insect to evade a strong
defensive response by the host. Such associations permit access to hosts, and, if these
weakened trees are in close spatial proximity, mountain pine beetle may then produce enough
offspring to strip-attack trees that are injured and/or colonized by other species of bark
beetles, or even mass-attack healthy neighboring trees (Eckberg et al. 1994, Logan et al.
1998). Clustering of trees strip-attacked by mountain pine beetle, evident in stands
undergoing a transition from endemic to incipient-epidemic levels, appeared to precipitate
mass attacks in neighbouring trees, as trees adjacent to successfully attacked hosts are likely
to become the foci of aggregation (Geiszler et al. 1980, Raffa and Berryman 1987, Eckberg et
al. 1994).
We note, however, that although the switch from trees with secondary colonization or
injury to healthy hosts is believed to be density-dependent (Carroll et al. 2006), spatial
analyses suggest that the behavioural shift in host colonization by mountain pine beetle is not
immediate, as the clustering process begins before epidemic levels have been reached.
Before mountain pine beetle reached the incipient-epidemic level, many trees harboring other
species of bark beetles and/or some form of injury were mass attacked. Prolonged endemic
behaviour may suggest genetic differences between beetles attacking weakened trees and
those attacking more vigourous hosts, as has been proposed in the spruce beetle D. ntfipennis
(Kirby) system (Wallin and Raffa 2004). Persistent endemic behaviour may also be the
product of varying phenotypes within the population whose tolerance for population densities

66

differ (Chitty 1958, Chitty and Phipps 1966). As the numbers of endemic beetles grow,
beetles phenotypically less tolerant of increasing densities may be prone to dispersal and seek
an alternative habitat such as healthy trees, leading to epidemic behaviour.
Spatial analyses within this study, in conjunction with the studies by Carroll et al.
(2006) and Smith (2008), lend additional support to an emerging theory of facilitation. It is
unknown how widespread such mechanisms of phase transitions may be in bark beetle
systems. In the southern pine beetle system, for example, the southern pine beetle D.
frontalis (Zimmerman) may benefit from associations with the secondaries /. avulsus
(Eichhoff) and /. calligraphus (Germar) that potentially help overcome tree resistance
(Wagner et al. 1985, Flamm et al. 1987).
This theoretical framework of shifting patterns of colonization by mountain pine
beetle from trees previously colonized by secondaries to fewer and fewer hosts with
secondaries and/or putative vigour impairing injuries, marked by the formation of clusters of
strip and mass attacks, suggests points of intervention that could be exploited for beetle
management. For example, the positive associations apparent between mountain pine beetle
and secondary bark beetles within weakened host trees suggest one reason why thinning
operations may be effective in preventing outbreaks (Mitchell et al. 1983, Larsson et al. 1983,
Raffa and Berryman 1986, Powell et al. 1998, Negron and Popp 2004). However, the benefits
of thinning are only realized if the risk of migration into the stand is low, as healthy trees may
still be overcome by large populations of mountain pine beetle migrating into a stand. This
phenomenon likely occurred in 2003 in stand E, for example. Furthermore, thinning

67

operations must remove material suitable for secondary bark beetle reproduction, as
population levels of secondaries may increase post-thinning (Kegley et al. 1997, Hindmarch
and Reid 2001), which may lead to further associations with endemic level mountain pine
beetle.

ACKNOWLEDGEMENTS
I thank B. S. Lindgren, F. R. McKee, and L. M. Poirier of the Forest Insect Research Group,
University of Northern British Columbia, for insight on this project. I also thank H. M. de la
Giroday for help with the implementation of the statistical platform for spatial modeling.
Funding for the project was provided by the Natural Sciences and Engineering Research
Council of Canada as well as Genome BC. Data collection for this project was funded by the
Government of Canada through the Mountain Pine Beetle Initiative, a six-year, $40 million
Program administered by Natural Resources Canada, Canadian Forest Service.

68

LITERATURE CITED
Abrams, P. A., R. D. Holt, and J. D. Roth. 1998. Apparent competition or apparent
mutualism? Shared predation when populations cycle. Ecology, 79: 201-212.
Akaike, H. 1973. A new look at the statistical model identification. IEEE Transactions on
Automatic Control 19: 716-723.
Amezaga, I., M. A. Rodriguez. 1998. Resource partitioning of four sympatric bark beetles
depending on swarming dates and tree species. Forest Ecology and Management,
109:127-135.
Amman, G. D. 1977. The role of mountain pine beetle in lodgepole pine ecosystems impact
on succession, pp. 3-18. In W. J. Mattson (ed.), Proceedings in life sciences the role of
arthropods in forest ecosystems. Springer-Verlag, New York, NY.
Amman, G. D. 1984. Mountain pine beetle (Coleoptera: Scolytidae) mortality in three types
of infestations. Environmental Entomology, 13: 184-191.
Amman G. D., and R. F. Schmitz. 1988. Mountain pine beetle: lodgepole pine interactions
and strategies for reducing tree losses. Ambio, 17: 62-68.
Aukema, B. H., A. L. Carroll, J. Zhu, K. F. Raffa, T. A. Sickley, and S. W. Taylor. 2006
Landscape level analysis of mountain pine beetle in British Columbia, Canada:
Spatiotemporal development and spatial synchrony within the present outbreak.
Ecography 29: 427-441.
Aukema, B. H., and K. F. Raffa. 2004. Does aggregation benefit bark beetles by diluting
predation? Links between a group-colonization strategy and the absence of multiple
predator effects. Ecological Entomology, 29: 129-138.
Ayres, B. D., M. P. Ayres, M. D. Abrahamson, and S. A Teale. 2001. Resource partitioning
and overlap in three sympatric species of Ips bark beetles (Coleoptera: Scolytidae).
Oecologia, 128: 443-453.
Ayres, M. P., R. T. Wilkens, J. J. Ruel, M. J. Lombardero, and E. Vallery. 2000. Nitrogen
budgets of phloem-feeding bark beetles with and without symbiotic fungi. Ecology,
81:2198-2210.
Baddeley, A., and R. Turner. 2000. Practical maximum pseudolikelihood for spatial point
patterns. Australian and New Zealand Journal of Statistics, 42: 283-322.

69

Baddeley, A., and R. Turner. 2005. Spatstat: an R package for analyzing spatial point patterns.
Journal of Statistical Software, 12: 1-42.
Bartos, D. L., and R. F. Schmitz. 1998. Characteristics of endemic level mountain pine beetle
populations in south central Wyoming. Research Paper RMRS-RP-13. United States
Department of Agriculture, Forest Service, Rocky Mountain Research Station, Ogden,
UT.
Batra, L. R. 1963. Ecology of ambrosia fungi and their dissemination by beetles. Transactions
of the Kansas Academy of Science, 66: 213-236.
Bergvinson, D. J., and J. H. Borden. 1991. Glyphosate-induced changes in the attack success
and development of the mountain pine beetle and impact of its natural enemies.
Entomologia Experimentalis et Applicata, 60: 203-212.
Berryman, A. A. 1972. Resistance of conifers to invasion by bark beetle-fungus associations.
Bioscience, 22: 598-602.
Bleiker, K. P., and D. L. Six. 2007. Dietary benefits of fungal associates to an eruptive
herbivore: potential implications of multiple associates on host population dynamics.
Environmental Entomology, 36: 1384-1396.
Boone, C. K., D. L. Six, and K. F. Raffa. 2008. The enemy of my enemy is still my enemy:
competitors add to predator load of a tree-killing bark beetle. Agricultural and Forest
Entomology, 10:411-421.
Byers, J. A. 1989. Chemical ecology of bark beetles. Experientia, 45: 271-283.
Carroll, A. L., B. H. Aukcma, K. F. Raffa, G. D. Smith, and B. S. Lindgren. 2006. Mountain
pine beetle outbreak development: the endemic-incipient transition. Mountain Pine
Beetle Initiative, Project 1.03 Working Paper, Natural Resources Canada, Canadian
Forest Service. 21p.
Carroll, A. L., S. W. Taylor, J. Regniere, and L. Safranyik. 2004. Effects of climate change on
range expansion by the mountain pine beetle in British Columbia, pp. 223-232. In
T. L. Shore, J. E. Brooks, and J. E. Stone (eds.), Mountain pine beetle symposium:
challenges and solutions. Natural Resources Canada, Canadian Forest Service, Pacific
Forestry Centre, Information Report BC-X-399, Victoria, British Columbia, Canada.
Chitty, D. 1958. Self-regulation of numbers through changes in viability. Cold Spring
Harbour Symposia of Quantitative Biology, 22: 277-280.

70

Chitty, D., and E. Phipps. 1966. Seasonal changes in survival in mixed populations of two
species of vole. Journal of Animal Ecology, 35: 313-331.
Cerezke, H. F. 1995. Egg gallery, brood production, and adult characteristics of mountain
pine beetle, Dendroctonus ponderosae Hopkins (Coleoptera: Scolytidae), in three pine
hosts. The Canadian Entomologist, 127: 955-965.
Christiansen, E., R. H. Waring, and A. A. Berryman. 1987. Resistance of conifers to bark
beetle attack: searching for general relationships. Forest Ecology and Management,
22: 89-106.
Cole, W. E., G. D. Amman. 1969. Mountain pine beetle infestations in relation to lodgepole
pine diameters. Research Note INT-95. United States Department of Agriculture,
Forest Service, Intermountain Forest and Range Experiment Station, Ogden, UT.
Cressie, N. A. C. 1991. Statistics for spatial data. John Wiley and Sons, New York, NY.
DeLeon, D. 1934. An annotated list of the parasites, predators, and other associated fauna of
the mountain pine beetle in western white pine and lodgepole pine. The Canadian
Entomologist, 66: 51-61.
Eckberg, T. B., J. M. Schmid, S. A. Mata, and J. E. Lundquist. 1994. Primary focus trees for
the mountain pine beetle in the Black Hills. Research Note RM- 531. United States
Department of Agriculture, Forest Service, Rocky Mountain Forest and Range
Experiment Station, Fort Collins, CO.
Flamm. R. O., T. L. Wagner, S. P. Cook, P. E. Pulley, R. N. Coulson, and T. M. McArdle.
1987. Host colonization by cohabiting Dendroctonus frontalis, Ips avulsits, and /.
calligmphus (Coleoptera: Scolytidae). Environmental Entomology, 16: 390-399.
Francke-Grossman, H. 1967. Ectosymbiosis in wood-inhabiting insects, pp. 141-203. In S.
M. Henry (ed.), Symbiosis. Academic Press, New York, NY.
Furniss, M. M., and S. J. Kegley. 2008. Biology of Dendroctonus murrayanae (Coleoptera:
Curculionidae: Scolytinae) in Idaho and Montana and Comparative Taxonomic Notes.
Annals of the Entomological Society of America, 101: 1010-1016.
Furniss, M. M., and J. A. Schenk. 1969. Sustained natural infestation by the mountain pine
beetle in seven new Pimis and Picea hosts. Journal of Economic Entomology, 62:
518-519.

71

Furniss, R. L., and V. M. Carolin. 1977. Western Forest Insects. United States Forest Service
Miscellaneous Publication 1339. United States Department of Agriculture, Forest
Service, Washington, D. C.
Geiszler D. R., V. F. Gallucci, and R. I. Gara. 1980. Modeling the dynamics of mountain pine
beetle aggregation in a lodgepole pine stand. Oecologia, 46: 244-253.
Goheen, D. J., and F. W. Cobb Jr. 1980. Infestation of ceratocystis wageneri-infected
ponderosa pines by bark beetles (Coleoptera: Scolytidae) in the central Sierra
Nevada. The Canadian Entomologist, 112: 725-730.
Hamel, D. R., and M. D. McGregor. 1976. Biological notes on the emergence of mountain
pine beetle and associates from lodgepole pine, Gallatin ranger district, Gallatin
National Forest, Montana 1975. Report 76-7. United States Department of Agriculture,
Forest Service, Northern Region, Missoula, MT.
Hindmarch, T. D., and M. L. Reid. 2001. Thinning of mature lodgepole pine stands increases
scolytid bark beetle abundance and diversity. Canadian Journal of Forest Research,
31: 1502-1512.
Hinds, T. E., L. R. Fuller, E. D. Lessard, and D. W. Johnson. 1984. Mountain pine beetle
infestation and Armillaria root disease of ponderosa pine in the Black Hills of South
Dakota. Technical Report R2-30. United States Department of Agriculture, Forest
Service, Timber, Forest Pest and Coopetative Forestry Management, Denver, CO.
Hodges, J. D., S. J. Barras, and J. K. Mauldin. 1968. Amino acids in inner bark of loblolly
pine, as affected by the southern pine beetle and associated microorganisms.
Canadian Journal of Botany, 46: 1467-1472.
lhaka, I., and R. Gentleman. 1996. R: a language for data analysis and graphics. Journal of
Computational and Graphical Statistics, 5: 299-314.
Kegley, S. J, R. L. Livingston, and K. E. Gibson. 1997. Pine engraver, Ips pini, in the western
United States. Forest insect and disease leaflet 122. USDA Forest Service,
Washington.
Kurz, W. A., C. C. Dymond, G. Stinson, G. J. Rampley, E. T. Neilson, A. L. Carroll, T.
Ebata, and L. Safranyik. 2008. Mountain pine beetle and forest carbon feedback to
climate change. Nature, 452: 987-990.
Larsson, S., R. Oren, R. H. Waring, and J. W. Barrett. 1983. Attacks of mountain pine beetle
as related to tree vigor of ponderosa pine. Forest Science, 29: 395-402.
72

Lessard, G., D. W. Johnson, T. E. Hinds, and W. H. Hoskins. 1985. Association of Armillaria
root disease with mountain pine beetle infestations on the Black Hills national forest,
South Dakota. Research Report 85-4. United States Department of Agriculture, Forest
Service, Timber Forest Pest Managemant, Fort Collins, CO.
Light, D. M., M. C. Birch, T. D. Paine. 1983. Laboratory study of intraspecific and
interspecific competition within and between two sympatric bark beetle species, Ips
pini and /. paraconfusus. Zeitschrift fur Angewandte Entomologie, 96: 233-241.
Logan, J. A., P. White, B. A. Bentz, and J. A. Powell. 1998. Model analysis of spatial patterns
in mountain pine beetle outbreaks. Theoretical Population Biology, 53: 236-255.
McCambridge, W. F., F. G. Hawksworth, C. B. Edminster, and J. G. Laut. 1982. Ponderosa
pine mortality resulting from a mountain pine beetle outbreak. Research Paper RM-23
United States Department of Agriculture, Forest Service, Rocky Mountain Forest and
Range Experiment Station, Fort Collins, CO.
Miller, D. R., and J. H. Borden. 1985. Life history and biology of Ips latidens (LeConte)
(Coleoptera: Scolytidae). Canadian Entomologist, 117: 859-871.
Mitchell, R. G., R. H. Waring, and G. B. Pitman. 1983. Thinning lodgepole pine increases
tree vigor and resistance to mountain pine beetle. Forest Science, 29: 204-211.
Negron, J. F., and J. B. Popp. 2004. Probability of ponderosa pine infestation by mountain
pine beetle in the Colorado Front Range. Forest Ecology and Management, 191: 1727.
Paine, T. D., K. F. Raffa and T. C. Harrington. 1997. Interactions among scolytid bark
beetles, their associated fungi, and live host conifers. Annual Review of Entomology,
42: 179-206.
Powell, J., J. Tarns, B. J. Bentz, and J. A. Logan. 1998. Theoretical analysis of "switching" in
a localized model for mountain pine beetle mass attack. Journal of Theoretical
Biology, 194: 49-63.
R Development Core Team. 2008. R: a language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria.
Raffa, K. F. 1991. Temporal and spatial disparities among bark beetles, predators, and
associates responding to synthetic bark beetle pheromones: Ips pini (Coleoptera:
Scolytidae) in Wisconsin. Environmental Entomology, 20: 1665-1679.

73

Raffa, K. F., B. H. Aukema, B. J. Bentz, A. L. Carroll, J. A. Hicke, M. G. Turner, and W. H.
Romme. 2008. Cross-scale drivers of natural disturbances prone to anthropogenic
amplification: the dynamics of bark beetle eruptions. Bioscience, 58: 501-517.
Raffa, K. F., and A. A. Berryman. 1983. The role of host plant resistance in the colonization
behavior and ecology of bark beetles (Coleoptera: Scolytidae). Ecological
Monographs, 53: 27-49.
Raffa, K. F., and A. A. Berryman. 1986. A mechanistic computer model of mountain pine
beele populations interacting with lodgepole pine stands and its implications for forest
managers. Forest Science, 32: 789-805.
Raffa, K. F., and A. A. Berryman. 1987. Interacting selective pressures in conifer-bark beetle
systems: a basis for reciprocal adaptations? The American Naturalist, 129: 234-262.
Rankin, L. J., and J. H. Borden. 1991. Competitive interactions between the mountain pine
beetle and the pine engraver in lodgepole pine. Canadian Journal of Forest Research,
21:1029-1036.
Rasmussen, L. 1974. Flight and attack behavior of mountain pine beetles in lodgepole pine of
northern Utah and southern Idaho. Research Note INT-180. United States Department
of Agriculture, Forest Service, Intermountain Research Station, Ogden, UT
Rasmussen, L. 1987. Mountain pine beetle selection of dwarf mistletoe and comandra blister
rust infected lodgepole pine. Research Note INT-367. United States Department of
Agriculture, Forest Service, Intermountain Research Station, Ogden, UT.
Ratchke, B. J. 1976. Competition and coexistence within a guild of herbivorous insects.
Ecology, 57: 76-87.
Reid, R. W. 1963. Biology of the mountain pine beetle, Dendroctomts monticolae Hopkins, in
the East Kootenay region of British Columbia. III. Interactions between the beetle and
its host, with emphasis on brood mortality and survival. The Canadian Entomologist,
95: 225-238.
Reid, R. W., H. S. Whitney, and J. A. Watson. 1967. Reactions of lodgepole pine to attack by
Dendroctonus ponderosae Hopkins and blue stain fungi. Canadian Journal of Botany,
45: 1115-1126.
Ripley, B. D. 1981. Spatial Statistics. J. Wiley, New York, NY.

74

Safranyik, L., and A. L. Carroll. 2006. The biology and epidemiology of the mountain pine
beetle in lodgepole pine forests, pp. 3-66. In L. Safranyik and B. Wilson (eds.), The
mountain pine beetle: a synthesis of its biology, management and impacts on
lodgepole pine. Pacific Forestry Centre, Canadian Forest Service, Natural Resources
Canada, Victoria, British Columbia.
Safranyik, L, T. L. Shore, A. L. Carroll, and D. A. Linton. 2004. Bark beetle (Coleoptera:
Scolytidae) diversity in spaced and unmanaged mature lodgepole pine (Pinaceae)
in southeastern British Columbia. Forest Ecology and Management, 200: 23-38.
Safranyik, L., T. L. Shore, D. A. Linton and L. Rankin. 1999. Effects of induced competitive
interactions with secondary bark beetle species on the establishment and survival of
mountain pine beetle broods in lodgepole pine. Information Report BC-X-384, Pacific
Forestry Centre, Canadian Forest Service, Natural Resources Canada, Victoria,
British Columbia. 40 p.
Safranyik, L., D. M. Shrimpton, and H. S. Whitney. 1974. Management of lodgepole pine to
reduce losses from the mountain pine beetle. Forestry Technical Report 1. Pacific
Forestry Centre, Canadian Forest Service, Natural Resources Canada, Victoria,
British Columbia. 29p.
Schenk, J. A., and D. M. Benjamin. 1969. Notes on the biology of Ips pini in central
Wisconsin jack pine forests. Annals of the Entomological Society of America, 62:
480-485.
Schlyter, F., and O. Anderbrant. 1993. Competition and niche separation between two bark
beetles: existence and mechanisms. Oikos, 68: 437-447.
Smith G. D. 2008. Maintenance of endemic Dendroctonus ponderosae populations through
interactions with a secondary bark beetle, Pseudips mexicanus. Master of Science
Thesis. University of Northern British Columbia. 121 p.
Smith G. D., A. L. Carroll, and B. S. Lindgren. 2009. Life history of a secondary bark beetle,
Pseudips mexicanus (Coleoptera: Curculionidae: Scolytinae), in lodgepole pine in
British Columbia. The Canadian Entomologist, 141: 56-69.
Smith, R. H., J. P. Cramer, and E. J. Carpender. 1981. New record of introduced hosts for the
mountain pine beetle in California. Research Note PSW-354. United States
Department of Agriculture, Forest Service, Pacific Southwest Forest & Range
Experiment Station, Berkeley, CA.
Stoyan, D., and A. Penttinen. 2000. Recent applications of point process methods in forestry
statistics. Statistical Science, 15: 61-78.

75

Thomson, A. J., and D. M. Shrimpton. 1984. Weather associated with the start of mountain
pine beetle Dendroctonus ponderosae outbreaks. Canadian Journal of Forest
Research, 14: 255-258.
Tkacz, B. M., and Schmitz, R. F. 1986. Association of an endemic mountain pine beetle
population with lodgepole pine infected by armillaria root disease in Utah. Research
Note INT-353. United States Department of Agriculture, Forest Service,
Intermountain Research Station, Ogden, UT.
Wagner, T. L., R. O. Flamm, and R. N. Coulson. 1985. Strategies for cohabitation among the
southern pine bark beetle species: comparisons of life-process biologies. General
Technical Report SO-56. United States Department of Agriculture, Forest Service,
Southeastern Forest Experiment Station, Asheville, NC.
Wallin, K. F., and K. F. Raffa. 2004. Feedback between individual host selection behavior
and population dynamics in an eruptive herbivore. Ecological Monographs, 74:
101-116.
Whitney, H. S. 1971. Association of Dendroctonus ponderosae (Coleoptera: Scolytidae) with
bluestain fungi and yeasts during brood development in lodgepole pine. Canadian
Entomologist, 103: 1495-1503.
Wood, S. L. 1982. The role of pheromones, kairomones, and allomones in the host selection
and colonization behavior of bark beetles. Annual Review of Entomology, 27:
411-446.

76

CHAPTER 4
Spatial associations of the bark beetles Dendroctonus murrayanae, Hylurgops spp., Ips pini,
Orthotomicus latidens and Pseudips mexicanus in lodgepole pine stands of
southern British Columbia
ABSTRACT
Non-eruptive bark beetles can be important agents of tree mortality that serve to thin
aging stands. Thinning of forests of lodgepole pine, Pinus contorta (Douglas ex Louden) is
of particular interest, because outbreaks of mountain pine beetle, Dendroctonus ponderosae
Hopkins, may be minimized in such stands. This study examines the spatial associations
between the non-eruptive bark beetles D. murrayanae (Hopkins), Hylurgops porosus
(LeConte), H. rugipennis (Mannerheim), Ips pini (Say), Orthotomicus latidens (LeConte),
and Pseudips mexicanus (Hopkins) in seven lodgepole pine stands in British Columbia from
2002 to 2005 while D. ponderosae was transitioning from endemic to incipient-epidemic
population phases. Trees colonized by O. latidens and P. mexicanus were located in close
proximity in all stands; in fact, these species were frequently colonizing the same host trees.
Trees colonized by Hylurgops spp. and P. mexicanus displayed similar spatial patterns, but
not as frequently as the former. The majority of trees colonized by these insects exhibited
some form of injury. Identifying the nature of interactions between bark beetles within the
bole-infesting assemblage may further our understanding of the propensity for stands to
harbour low density populations of mountain pine beetle and ultimately undergo an outbreak.
Key words: competition; niche partitioning; pine engravers; forest management

77

INTRODUCTION
Bark beetles (Coleoptera: Curculionidae, Scolytinae) are important disturbance agents
in forest ecosystems. Eruptive species may undergo intermittent population explosions
resulting in landscape-level mortality (Raffa et al. 2008). The vast majority of bark beetle
species, however, do not undergo such dramatic population changes, but instead subsist in
weakened, dying, and dead trees (Rudinsky 1962, Wood 1982a). Although the number of
vigour impaired or unthrifty trees containing phloem suitable for reproduction within a stand
is often limited (Berryman 1973, Anderbrant et al. 1985), increased resource availability due
to stress events, such as drought, may facilitate rapid growth in these non-eruptive bark beetle
populations (Hopping and Mathers 1945, Rudinsky 1962, Mattson and Haack 1987, Kelsey
and Joseph 2001, Berg et al. 2006). These "secondary" bark beetles are vital components of
forest ecosystems, beneficial to forest succession, and essential to the perpetuation of
vigorous trees (Lundquist 1995, Jones et al. 1997).
A diverse number of bark beetles have been identified inhabiting lodgepole pine,
Finns contorta (Douglas ex Louden), forests within western North America (Bright 1976,
Safranyik et al. 2004, Carroll et al. 2006). Bark beetle species that may be found in British
Columbia, Canada, include Dendroctonus murrayanae (Hopkins), Hylurgops porosus
(LeConte), H. rugipennis (Mannerheim), Ips pint (Say), Orthotomicus latidens (LeContc),
and Pseudips mexicamis (Hopkins). Each of these insects, like mountain pine beetle,
Dendroctonus ponderosae Hopkins, during its endemic population phases, is often associated
with the boles of weakened pine hosts. Collectively, they are termed the bole-infesting bark

78

beetle assemblage (Safranyik and Carroll 2006).
Thinning is considered to be one of the few effective management options to reduce
the growth rate and outbreak potential of eruptive species of bark beetles (e.g., see review by
Fettig et al. 2007). There is a wide body of empirical evidence in support of thinning to
reduce outbreak extent in the mountain pine beetle system in both stands of lodgepole and
ponderosa pine, Pinus ponderosa (Douglas ex Lawson) (Sartwell and Stevens 1975, Mitchell
et al. 1983, Larsson et al. 1983, Raffa and Berryman 1986, McGregor et al. 1987, Negron and
Popp 2004, Whitehead and Russo 2005). The mechanisms by which thinning reduces bark
beetle populations are multi-causal, and may include changes in microclimate within a stand
(Amman et al. 1988, Hindmarch and Reid 2001), as well as increases in host vigour within
the remaining trees released from competition (Mitchell et al. 1983, Raffa and Berryman
1986). It thus stands to reason that growing populations of bole-infesting bark beetles may
influence the development of endemic and incipient-epidemic levels of mountain pine beetle
not only through within-tree competition for resources, but also through natural stand
attenuation or thinning.
In this chapter, I examined the spatiotemporal dynamics of the non-eruptive
assemblage of bark beetles in the years preceding an outbreak of mountain pine beetle in
several stands of lodgepole pine in British Columbia, Canada. Many of these species
reproduce in phloem not utilized by the eruptive species of bark beetles during an outbreak
(Safranyik et al. 1974, Furniss and Carolin 1977, Wood 1982b), such that subsequent postoutbreak numbers of these insects may kill small-diameter trees (Furniss and Carolin 1977,

79

Paine et al. 1997, Steed and Wagner 2008). Less is known however, about population
dynamics of these insects in years preceding a large outbreak of mountain pine beetle. The
relationship between secondary bark beetles and mountain pine beetle is complex, as activity
by non-eruptive bark beetles may facilitate populations of endemic mountain pine beetle
rather than putatively increase stand resistance (Carroll et al. 2006, Smith et al. 2009; Chapter
3). To gain an understanding of how these insects potentially mediate dynamics of
outbreaking species, we first need to understand the behaviour within their assemblage. To
that end, this chapter examined the spatial interactions of the bole-infesting non-eruptive bark
beetle assemblage prior to the eruption of populations of mountain pine beetle at a withinstand level. I examined whether the guild shares a comon resource, and, if so, exhibits a
consistent sequence of activity. I also examined whether there is an association between
putative vigour-impairing injuries and colonization. Answers to these questions of
predisposing factors and colonization sequences may shed light on the mechanisms by which
bark beetles erupt (Aukema et al. 2010, Boone et al. 2010 in press), and, by extenstion, may
suggest new tactics for their management.
METHODS
Seven field sites were established in southern British Columbia and monitored for
bark beetle activity by exhaustively censusing each tree. Trees colonized by bark beetles
were then visually evaluated for injury such as a broken or forked top, a scarred trunk, a thin
crown, an infection of dwarf mistletoe (Arceuthobium spp.), an attack by mountain pine
beetle greater than ten years prior, or some other form of injury/suppression. For a detailed

80

description, see the Methods section of Chapter 2. Maps of all trees colonized by secondary
bark beetles were created using the 'spatstat' package v.1.13-3 in R v.2.6.2 (Ihaka and
Gentleman 1996, Baddeley and Turner 2005, R Development Core Team 2008).
Colonization of trees by secondary bark beetles was analysed by year and species. For all
stands, with the exception of stand D, the secondaries included O. latidens, P. mexicanus, I.
pini, H. porosus, and H. rugipennis. Hylurgops porosus and H. rugipennis are collectively
termed Hylurgops spp. Ips pini were not analyzed in stand D due to low numbers.
Dendroctonus murrayanae was analyzed in stands D and G, where sufficient numbers
existed. Analyses were conducted only when there were at least two trees colonized by each
species of secondary bark beetle in a given year.
Spatial point process models were used to evaluate the spatial relationships between
secondary colonization events within each stand. New techniques in spatial point process
modeling are quite powerful for discerning potential relationships between species or other
biotic and abiotic factors that may otherwise go undetected (Stoyan and Penttinen 2000).
Each spatial point process model incorporated previous years of colonization as well as
colonization occurring in the same year. The response variable for each model, a density (A),
was measured as the number of trees bearing colonization of a given species per square meter
in a given year. For example, a spatially-explicit estimated density of O. latidens colonization
was measured as the number of trees bearing O. latidens attack per square meter in any given
year. In a model with O. latidens as a response, covariates could include the locations of
colonization by O. latidens and all other secondaries in the preceding year, as well as all

81

other secondaries in that same year. All covariates were converted to spatially-explicit
density surfaces (i.e., the average number of trees colonized in a given area in a given stand)
prior to fitting using a Gaussian kernel density smoother (Cressie 1991, Baddeley and Turner
2000). Parameters in these spatial point process regression models were estimated using
maximum pseudolikelihood methods. The significance of individual variables were judged
by statistical comparison to a homogenous model, i.e., one estimating only an intercept or a
constant intensity of secondary bark beetle colonization across the stand, by examining the
change in deviance relative to a % 2 reference distribution. Models were compared using
Akaike's Information Criterion (AIC), and models with the lowest AIC values were judged to
fit best (Akaike 1973).
RESULTS
The colonization of lodgpepole pine trees by secondary bark beetles within study
stands was generally quite low. Between 0.2% (stand C), and 3% (stand D) of the available
host trees were colonized by secondary bark beetles over the seven year study (see Appendix
A Table A.I). These estimates were calculated based on the number of secondary bark beetle
attacks, the size of each stand in hectares, the average density of trees per hectare, and the
percent of lodgepole pine within each stand. The number of injured trees was much higher,
however. The mean injury rate of trees across all stands was 48%. Seventy-four percent of
colonization by secondary bark beetles across all stands occurred in trees possessing at least
one putative vigour-impairing injury (Table 4.1), such as a broken or forked top, a scarred
trunk, a thin crown, an infection of dwarf mistletoe {Arceuthobiuin spp.), an attack by

82

mountain pine beetle greater than ten years prior, or some other form of injury or
suppression.
Patterns and sequences of colonization between secondary bark beetle species were
generally consistent among all seven stands. To simplify data presentation, results of
analyses are presented primarily from stand A (1999-2005). Other than some information on
D. murrayanae in stand D, which is also reported, analyses of the remaining stands for all
species are found in Appendix D.
Pseudips mexicaniis and O. latidens frequently colonized the same host trees in the
same year (Tables 4.2 and 4.3). A visual representation of this cluster pattern is provided in
Fig. 4.1, where approximately 0.6% and 0.4% of the trees in the stand were colonized by P.
mexicaniis and O. latidens, respectively, in 2002 (Fig. 4.1 A). Orthotomicus latidens also
colonized trees inhabited by P. mexicaniis in previous years, estimated to be approximately
78 trees throughout the stand in 2002 (Fig. 4.IB). Pseudips mexicaniis also re-attacked trees
quite frequently (Fig. 4.2).
The locations of trees colonized by O. latidens in a previous year were a good
predictor of the presence of O. latidens colonization in a subsequent year, indicative of reattack or colonization of neighbouring pine hosts (Table 4.2, and see Appendix D: Tables
D.3-D.5). However, the best inference on the locations of trees colonized by O. latidens in a
given year were the locations of trees previously attacked by P. mexicaniis as judged by lower
AIC values for these models (Table 4.2).
Pseudips mexicaniis had a tendency to inhabit hosts colonized either previously or

83

concurrently by Hylurgops spp. particularly in the earlier years of study in each stand (Tables
4.3 and 4.4, Fig. 4.3A, and see Appendix D: Tables D.6-D.18). Trees colonized by
Hylurgops spp. were near trees colonized by O. latidens in the same year and/or in future
years (Tables 4.2 and 4.4, Fig 4.3B). However, Hylurgops spp. were generally not found in
trees that had been colonized by O. latidens previously. Rather, O. latidens appeared to
colonize hosts already inhabited by Hylurgops.
Positive associations between trees colonized by /. pini and those colonized by any
other secondary species under investigation were apparent, and are detailed in Appendix D
(Tables D.18-D.22). The locations of trees colonized by /. pini were often proximate to those
colonized by P. mexicanus, such that the two were frequently found utilizing the same host
trees (Table 4.5, Fig. 4.4). There were, however, solitary instances when the location of/.
pini colonization was negatively associated with colonization by each of P. mexicanus, O.
latidens, and Hylurgops spp. (Appendix D: Tables D.10, D.18, D.20). However, spatial
analyses at the stand-level across years provided little evidence of inhibition between /. pini
and other secondary bark beetles.
D. murrayanae colonization was present in low numbers in most stands, and was
incorporated into spatial analyses in stands D and G only. Spatial analyses indicated that the
locations of D. murrayanae could not be consistently explained by knowing the locations of
other species of bark beetles, although D. murrayanae would occasionally colonize trees
containing P. mexicanus (Appendix D: Table D.9), Hylurgops spp. (Appendix D: Table D.18)
and /. pini (Appendix D: Table D.22). Once D. murrayanae had colonized a tree, it appeared

84

to re-attack the same tree or neighbouring trees (Table 4.6).

85

Table 4.1: Number of lodgepole pine (Pinus contorta) trees colonized by one or more species
of secondary bark beetle (Dendroctonus murrayanae, Hylurgops spp., Ips pini, Orthotomicus
latidens, and/or Pseudips mexicanus) and, within those same trees, the number bearing
injuries within seven stands in southern British Columbia.
No. Trees
Year
Stand
1999
2002
2005
2000
2001
2003
2004
Secondaries
193
96
A
9
20
157
263
130
64
9
155
256
138
Injured
18
88
B

C

D

E

F

G

Secondaries
Injured

_

27
25

123
116

196
187

136
107

_

_

-

-

Secondaries
Injured

-

11
10

43
40

36
34

13
8

14
5

11
0

Secondaries
Injured

_

_

_

-

-

-

4
4

14
9

17
7

3
1

Secondaries
Injured

_

25
23

91
71

235
168

_

_

-

2
0

-

-

Secondaries
Injured

-

_

_

-

-

-

26
18

59
38

171
95

94
59

Secondaries
Injured

-

-

-

-

1
1

19
9

137
75

314
158

160
73

-

86

Table 4.2: Best explanatory models for the location of Orthotomicus latidens colonization
from 2001 to 2005 in lodgepole pine of stand A in southern British Columbia. The line in
bold represents an intercept-only model; i.e., modeling a constant density of insects across
the stand. Subsequent lines reflect whether the location of each listed insect and year
provides inference on the location of the insect studied relative to this constant density. A
positive estimate for a slope reflects positive spatial association, while a negative estimate
reflects spatial inhibition at a between-tree scale. The response variable for each equation is
log(A), where A is the density estimate of trees colonized by a given species per square meter.
For example, the estimated density of O. latidens colonization in 2002 in locations where
Pseudips mexicanus colonized trees at a rate of 0.0005/m2 or 5 trees/ha. is e X p ( 9 7 1 + 2324x000(b)
= 0.0002 or 2 trees/ha. For each, year significant models are listed in order of best fit.
2
Insect
Year
Intercept
Slope
P-value
AIC

x

Est.
O. latidens
O. latidens

2001
2000

-9.14
-9.42

SE
0.22
0.26

O. latidens
P. mexicanus
H. spp.
P. mexicanus
H. spp.

2002
2002
2002
2001
2001

-7.85
-9.71
-8.70
-9.22
-8.50

0.11
0.34
0.21
0.32
0.24

O. latidens
P. mexicanus
P. mexicanus
H. spp.

2003
2003
2002
2003

-8.12
-9.78
-9.07
-8.54

0.13
0.53
0.35
0.21

O. latidens
P. mexicanus
H. spp.
H. spp.

2004
2004
2003
2004

-8.72
-10.52
-9.54
-9.36

0.18
0.48
0.30
0.28

O. latidens
2005
P. mexicanus 2005
P. mexicanus 2004

-8.55
-10.00
-9.57

0.16
0.37
0.39

Est.
13531

2324
4784
2916
2548

3027
1286
3357

3495
11489
8346

4509
2152

87

SE
3943

351
797
555
737

881
399
1091

714
2440
2268

800
663

8.43

44.45
29.88
26.26
11.57

13.38
10.20
8.09

23.09
19.44
11.14

23.09
9.77

0.004

427.85
421.43

<0.0001
<0.0001
<0.0001
0.0007

1347.67
1305.23
1319.80
1323.43
1338.11

<0.001
<0.001
0.004

1060.31
1048.94
1052.13
1054.23

<0.000l
<0.0001
0.001

623.96
602.87
606.52
614.83

<0.0001
0.002

727.51
699.84
719.75

Table 4.3: Best explanatory models for the location of Pseiidips mexicanus colonization from
2000 to 2005 in lodgepole pine of stand A in southern British Columbia. The line in bold
represents the null model for each year; i.e., no explanatory variable (reflected by the absence
of an estimated slope) thus modeling a constant density of insects across the stand.
Subsequent lines reflect whether the location of each listed insect and year provides inference
on the location of the insect studied relative to this constant density. A positive estimate for a
slope reflects positive spatial association, while a negative estimate reflects spatial inhibition
at a between-tree scale. The response variable for each equation is log(A), where A is the
density estimate of trees colonized by a given species per square meter. For each year
significant models are listed in order of best fit.
2
Insect
P-value
Year
Intercept
Slope
AIC

x

SE
0.30
0.44

Est.

SE

P. mexicanus
H. spp.

Est.
2000 -9.79
2000 -10.43

12882

4193

7.30

0.01

239.29
233.99

P. mexicanus
H. spp.
P. mexicanus
P. mexicanus

2001
2001
2000
2001

-7.81
-8.59
-8.03
-8.07

0.11
0.24
0.14
0.17

3019
3180
3587

720
1087
1539

17.04
7.06
5.04

<0.0001
0.01
0.02

1394.67
1379.65
1389.62
1391.64

P. mexicanus
O. latidens
H. spp.
P. mexicanus
H. spp.
I. pini

2002
2002
2002
2001
2001
2002

-7.33
-8.40
-8.13
-8.43
-7.90
-7.64

0.09
0.20
0.18
0.24
0.18
0.14

<0.0001
<0.0001
<0.0001
<0.0001
0.003

2134.93
2091.58
2103.83
2108.67
2121.66
2128.03

P. mexicanus
O. latidens
H. spp.

2003
2003
2003

-7.63
-8.33
-7.90

0.10
0.25
0.16

2110
2247

648
923

10.33
5.31

0.001
0.02

1641.69
1633.38
1638.40

P. mexicanus
O. latidens
H. spp.
H. spp.
O. latidens

2004
2004
2003
2004
2003

-7.81
-8.45
-8.29
-8.18
-8.45

0.11
0.19
0.17
0.17
0.27

3027
7521
9602
1915

595
1662
2994
711

21.25
18.23
9.43
7.06

<0.0001
<0.0001
0.002
0.01

1394.67
1375.43
1378.45
1387.26
1389.63

P. mexicanus
O. latidens
P. mexicanus
H. spp.

2005
2005
2004
2005

-8.31
-9.70
-9.25
-8.53

0.14
0.33
0.34
0.17

<0.0001
0.001
0.01

896.01
863.10
887.73
890.20

2242
6736
2365
2270
2418

341
1152
434
571
769

5429
908
1975
594
13711 4420

88

45.38
33.13
28.29
15.30
8.93

34.92
10.29
7.82

Table 4.4: Best explanatory models for the location of Hylurgops spp. colonization from 2001
to 2004 in lodgepole pine of stand A in southern British Columbia. The line in bold
represents the null model for each year; i.e., no explanatory variable (reflected by the absence
of an estimated slope) thus modeling a constant density of insects across the stand.
Subsequent lines reflect whether the location of each listed insect and year provides inference
on the location of the insect studied relative to this constant density. A positive estimate for a
slope reflects positive spatial association, while a negative estimate reflects spatial inhibition
at a between-tree scale. The response variable for each equation is log(A), where A is the
density estimate of trees colonized by a given species per square meter. For each year
significant models are listed in order of best fit.
1
Intercept
Insect
Year
Slope
P-value
AIC

x

SE
0.18
0.48
0.49
0.23

Est.

SE

9137
2923
4095

1727
857
1567

H. spp.
0. latidens
P. mexicanus
P. mexicanus

2001
2001
2001
2000

Est.
-8.72
-10.64
-10.09
-9.01

H. spp.
P. mexicanus
O. latidens
H. spp.
P. mexicanus
0. latidens

2002
2002
2002
2001
2001
2001

-9.24
-11.23
-10.67
-10.42
-10.66
-9.75

0.23
0.70
0.55
0.55
0.64
0.35

2479
2873
6058
3020
3582

710
819
2231
1111
1461

H. spp.
P. mexicanus
0. latidens
0. latidens
P. mexicanus

2003
2002
2003
2002
2003

-9.99
-13.37
-12.22
-11.58
-12.38

0.33
1.25
1.02
0.81
1.52

3857
5886
3142
4263

1126
2135
1186
2430

H. spp.
0. latidens
H. spp.

2004 -10.39
2003 -12.96
2003 -11.47

0.41
1.32
0.74

6610
13870

89

2667
5534

<0.0001
0.001
0.02

623.96
598.39
614.90
620.59

12.46
11.53
7.05
7.03
4.98

0.0004
0.001
0.01
0.01
0.03

391.10
380.64
381.57
386.05
386.07
388.12

13.56
7.84
6.61
3.73

0.0002
0.01
0.01
0.05

199.76
188.20
193.92
195.15
198.03

0.01
0.02

138.70
134.19
135.13

27.57
11.06
5.38

6.52
5.58

Table 4.5: Best explanatory models for the location of Ips pini colonization from 2001 to
2003 in lodgepole pine of stand A in southern British Columbia. The line in bold represents
the null model for each year; i.e., no explanatory variable (reflected by the absence of an
estimated slope) thus modeling a constant density of insects across the stand. Subsequent
lines reflect whether the location of each listed insect and year provides inference on the
location of the insect studied relative to this constant density. A positive estimate for a slope
reflects positive spatial association, while a negative estimate reflects spatial inhibition at a
between-tree scale. The response variable for each equation is log(A), where A is the density
estimate of trees colonized by a given species per square meter. For each year significant
models are listed in order of best fit.
Intercept
Est. SE
-9.70 0.29
-10.99 0.79

Slope
Est.
SE

x1

P-value

AIC

2763

1399

3.70

0.05

258.77
257.07

2002
2002
2002
2001

-9.14
-10.55
-10.31
-9.60

0.22
0.52
0.60
0.34

2829
1554
5964

777
661
2795

12.39
5.43
4.09

0.0004
0.02
0.04

427.85
417.46
424.42
425.77

2003
I. pini
P. mexicanus 2003
2003
0. latidens

-9.54
-11.65
-10.78

0.27
1.17
0.71

3790
3525

1890
1672

4.75
4.35

0.03
0.04

297.25
294.51
294.91

Insect

Year

/. pini
2001
P. mexicanus 2001
I. pini
O. latidens
P. mexicanus
I. pini

90

Table 4.6: Best explanatory models for the location of Dendroctonus murrayanae attack from
2003 and 2005 in lodgepole pine of stand D in southern British Columbia. The line in bold
represents the null model for each year; i.e., no explanatory variable (reflected by the absence
of an estimated slope) thus modeling a constant density of insects across the stand.
Subsequent lines reflect whether the location of each listed insect and year provides inference
on the location of the insect studied relative to this constant density. A positive estimate for a
slope reflects positive spatial association, while a negative estimate reflects spatial inhibition
at a between-tree scale. The response variable for each equation is log(A), where A is the
density estimate of trees colonized by a given species per square meter. For each year
significant models are listed in order of best fit.
Insect

Year

D. murrayanae
2003
D. murrayanae 2002

Intercept
Est.
SE
-8.69 0.23
-9.65 0.54

D. murrayanae 2005
D. murrayanae 2004

-8.45
-9.26

0.20
0.46

Slope
Est. SE
8976

3232

91

3484

1479

%

4.56

4.03

P-value

AIC

0.03

370.12
367.33

0.05

455.78
453.46

x = P. mexicanus 2002
o = 0. latidens 2002

##

340m

0

<h

6

\
x

770m
B

x = P. mexicanus 2001
o = 0 latidens 2002

o

a

340m
o

8 "

770m
Figure 4.1: Locations of secondary bark beetle colonization in lodgcpole pine of stand A in
southern British Columbia. A) Trees colonized by Pseudips mexicanus and Orthotomicus
latidens in 2002 B) Trees colonized by P. mexicanus in 2001 and O. latidens in 2002
comprise approximately 0.4% of the 19,500 lodgepole pine trees in the stand respectively.

92

x=P. mexicanus 2001
o = P. mexicanus 2002

o „ a

340m

0

0

8
x»
0
«
X o o

„o

3

'

o "<
*

°

c

r,o

U

xo°'J

,,?

*

x

x

770m
Figure 4.2: Locations of trees colonized by Pseudips mexicanus in lodgepole pine of stand A
in southern British Columbia. Colonizations by P. mexicanus in 2001 and 2002 comprise
approximately 0.4% and 0.6% of the 19,500 lodgepole pine trees in the stand respectively.

93

x = H. spp. 2002
o = P. mexicamis 2002

00

340m

0
0

JO
0

X

0 °
0

0

0
0

0 o

0

00

0

o
0

770m
B

340m

x = H. spp 2002
o = 0. latidens 2002

§s°

«w

770m
Figure 4.3: Locations of secondary bark beetle colonization in lodgepole pine of stand A in
southern British Columbia. A) Trees colonized by Hylurgops spp. and Pseudips mexicamis in
2001 comprise approximately 0.2% and 0.4% of the 19,500 logepole pine trees in the stand
respectively. B) Trees colonized by H. spp. and Orthotomicus latidens in 2002 comprise
approximately 0.1% and 0.4% of the 19,500 lodgepole pine trees in the stand respectively.

94

\=P. mexicanus 2003
o = I. pini 2003

340m
<\

770m
Figure 4.4: Location of trees colonized by Pseudips mexicanus and Ips pini in lodgepole pine
of stand A in southern British Columbia. Colonizations by P. mexicanus and /. pini in 2003
comprise approximately 0.45% and 0.07% of the 19,500 lodgepole pine trees in the stand
respectively.

95

DISCUSSION
This study suggests that colonization dynamics of secondary bark beetles in stands of
lodgepole pine are not entirely random, but rather follow a stereotypic sequence of events and
associations. The frequent within or between-tree associations between Hylurgops spp. and
P. mexicanus, Hylurgops spp. and O. latidens, and /. pini and P. mexicanus from the endemic
to the incipient-epidemic stages of population development of mountain pine beetle suggests
that sufficient resource partitioning must occur to minimize interspecific competition and
promote a robust assemblage of bark beetles (Birch et al. 1980, Byers 1989).
Temporal partitioning by secondary bark beetles is one effective means of limiting
interspecific competition (Paine et al. 1981, Amezaga and Rodriguez 1998, Safranyik et al.
2004). However, temporal analyses (Chapter 2) suggested emergence, flight, and host
colonization by these species of bark beetles overlap substantially. Hylurgops spp. emerge
and colonize hosts prior to, or in quick succession with, P. mexicanus. The tendency for O.
latidens to infest trees already occupied by P. mexicanus or Hylurgops spp. may be at least
partially explained by its prolonged emergence through the time of study in each year, as well
as its ubiquitous occurence in tissues in more advanced stages of deterioration (Miller and
Borden 1985). For example, while O. latidens often reproduced in a host in the same year or
in the year following colonization by Hylurgops spp. or P. mexicanus, O. latidens galleries do
not overlap those of P. mexicanus, as colonization by P. mexicanus often occurs below O.
latidens and /. pini galleries (A. Carroll, pers. comm., Smith et al. 2009). Although P.
mexicanus may be found at greater heights and into the branches of trees in areas of

96

California (Fox et al. 1990) and Mexico (Schwerdtfeger 1956), attacks in lodgepole pine trees
within British Columbia generally do not occur above one metre (Smith et al. 2009).
Hylurgops porosus and H. rugipennis utilize the lower bole of various pine hosts at or below
the root collar (Wood 1982b).
Avoidance of occupied hosts is a simple way to reduce interspecific competition. Ips
pini ordinarily avoids trees inhabited by O. latidens (Poland 1997) or P. mexicanus (Poland
1997, Savoie 1998). We found, for example, that I. pini colonization was located on the
westernmost side of stand E, while O. latidens and P. mexicanus inhabited trees on the
eastern half. However, the new finding of positive spatial associations between I. pini and P.
mexicanus colonization in our study may be attributed to the extensive number of trees
surveyed. It is possible that /. pini, with the development of multiple generations within a
year (Furniss and Carolin 1977), may have attacked hosts after P. mexicanus colonization had
occurred. Ips pini often predominates phloem resources in the hosts it colonizes. Any
remaining phloem lower down on the bole near the root collar, however, may be utilized by P.
mexicanus (E. Teen, pers. comm.).
The majority of colonization occurred in trees with putative vigour-impairing injuries.
The declining rate of trees with injuries colonized by non-eruptive bark beetles from 2003 to
2005 might be explained by mountain pine beetle activity. In this study, growing populations
of mountain pine beetle shifted from injured and/or weakened trees to uninjured, vigorous
hosts (Chapter 3; Carroll et al. 2006), and, after completing their life-cycle, left behind a
source of previously uninjured hosts available for colonization (Safranyik et al. 1974, Furniss

97

and Carolin 1977, Wood 1982b, Paine et al. 1997, Steed and Wagner 2008). For stands D and
G, where populations of mountain pine beetle did not exit an endemic phase, the large
number of uninjured trees colonized by non-eruptive species could likely be attributed to P.
mexicanus and O. latidens attack on small diameter, yet otherwise healthy, trees (Furniss and
Carolin 1977, Paine et al. 1997). Stand D also contained the oldest trees, with an average age
of nearly 150 years (Carroll et al. 2006). An infestation of D. murrayanae in uninjured,
senescent hosts within this stand also likely contributed to a large number of the trees lacking
injuries that were colonized by non-eruptive species of bark beetles.
The colonization behaviour of robust assemblages of bark beetles may naturally thin
stands as weaker trees are killed. Trees not suffering mortality by bark beetles benefit from
reduced competition for light, water, and nutrients (Oliver and Larson 1996). Lodgepole pine
stands may outgrow their susceptibilty to localized eruptions of mountain pine beetle if the
majority of weakened hosts have been naturally thinned by the activity of secondary bark
beetles. In this study, 74% of trees attacked by secondary bark beetles displayed an injury or
some form of disease. Thinning activity may extinguish or severely limit reservoirs of
endemic mountain pine beetle by eliminating their hosts. This could explain why some
stands upwards of 160 to 170 years of age have never suffered serious depletion from more
aggressive species of bark beetles (Hopping and Mathers 1945).
Resource sharing between bark beetles within the bole-infesting assemblage may
enable or prolong localized eruptions of typically non-eruptive species following an epidemic
of mountain pine beetle and contribute to the mortality of smaller, but otherwise healthy

98

trees. Future work should focus on the nature of interactions between these species.
Competition experiments, for example, between varying densities of bark beetles in the
complete bole-infesting assemblage may lend valuable information concerning the positive
spatial associations found here. A greater understanding of the extent of resource partitioning
or competition at a within-tree level may enhance the projection of mortality to small trees
following outbreaks of mountain pine beetle. Further research is also required to fully
understand the interactions of non-eruptive species before they may be utilized as natural
agents useful in stand management. However, the existence of stands in which bark beetles
have thinned trees to a level such that they are essentially immune to localized eruptions
suggests that natural approaches to future management have promise.

ACKNOWLEDGEMENTS
I thank B. S. Lindgrcn, F. R. McKee, and L. M. Poirier of the Forest Insect Research Group,
University of Northern British Columbia, for their insight on this project. I also thank H. M.
de la Giroday for help with the implementation of the statistical platform for spatial
modeling. Funding for this project was provided by the Natural Sciences and Engineering
Research Council of Canada as well as Genome BC. Data collection for this project was
funded by the Government of Canada through the Mountain Pine Beetle Initiative, a six-year,
$40 million Program administered by Natural Resources Canada, Canadian Forest Service.

99

LITERATURE CITED
Akaike, H. 1973. A new look at the statistical model identification. IEEE Transactions on
Automatic Control 19: 716-723.
Amezaga, I., and M. A. Rodriguez. 1998. Resource partitioning of four sympatric bark
beetles depending on swarming dates and tree species. Forest Ecology and
Management, 109:127-135.
Amman, G. D., and R. F. Schmitz. 1988. Mountain pine beetle - lodgepole pine interactions
and strategies for reducing tree losses. Ambio, 17: 62-68.
Anderbrant, O., F. Schlyter, and G. Birgersson. 1985. Intraspecific competition affecting
parents and offspring in the bark beetle Ips typographus. Oikos, 45: 89-98.
Aukema, B. H., J. Zhu, J. M0ller, J. G. Rasmussen, and K. F. Raffa. 2010. Predisposition to
bark beetle attack by root herbivores and associated pathogens: Roles in forest decline,
gap formation, and persistence of endemic bark beetle populations. Forest Ecology
and Management, 259: 374-382.
Baddeley, A., and R. Turner. 2000. Practical maximum pseudolikelihood for spatial point
patterns. Australian and New Zealand Journal of Statistics, 42: 283-322.
Baddeley, A., and R. Turner. 2005. Spatstat: an R package for analyzing spatial point patterns.
Journal of Statistical Software, 12: 1-42.
Berg, E. E., J. D. Henry, C. L. Fastic, A. D. De Voider, and S. M. Matsuoka. 2006. Spruce
beetle outbreaks on the Kenai Peninsula, Alaska, and Kluane National Park and
Reserve, Yukon Territory: relationship to summer temperatures and regional
differences in disturbance regimes. Forest Ecology and Management, 227: 219-232.
Berryman, A. A. 1973. Population dynamics of the fir engraver, Scolytus ventralis
(Coleoptera: Scolytidae). I. Analysis of population behavior and survival from 1964
to 1971. The Canadian Entomologist, 105: 1465-1488.
Birch, M. C , P. Svihra, T D. Paine, and J. C. Miller. 1980. Influence of chemically mediated
behavior on host tree colonization by four cohabiting species of bark beetles. Journal
of Chemical Ecology, 6: 395-414.
Boone, C. K., B. H. Aukema, J. Bohlmann, A. L. Carroll, and K. F. Raffa. 2010. Efficacy of
plant defenses varies with herbivore population density. Canadian Journal of Forest
Research, Accepted pending revision.
100

Bright, D. E., Jr. 1976. The bark beetles of Canada and Alaska. Coleoptera: Scolytidae.
Canada Department of Agriculture, Biosystematic Research Institute Research
Branch, Ottawa, Ontario, Publication 1576. 241p.
Byers, J. A. 1989. Chemical ecology of bark beetles. Experientia, 45: 271-283.
Carroll, A. L., B. H. Aukema, K. F. Raffa, G. D. Smith, and B. S. Lindgren. 2006. Mountain
pine beetle outbreak development: the endemic-incipient transition. Mountain Pine
Beetle Initiative, Project 1.03 Working Paper, Natural Resources Canada, Canadian
Forest Service. 21p.
Cressie, N. A. C. 1991. Statistics for spatial data. John Wiley and Sons, New York, NY.
Fettig, C. J., K. D. Klepzig, R. F. Billings, A. S. Munson, T. E. Nebeker, J. F. Negron, and J.
T. Nowak. 2007. The effectiveness of vegetation management practices for prevention
and control of bark beetle infestations in coniferous forests of the western and
southern United States . Forest Ecology and Management, 238: 24-53 .
Fox, J. W., D. L. Wood, and C. S. Koehler. 1990. Distribution and abundance of engraver
beetles (Scolytidae: Ips species) on Monterey pines infected with pitch canker. The
Canadian Entomologist, 122: 1157-1166.
Furniss, R. L., and V. M. Carolin. 1977. Western Forest Insects. United States Forest Service
Miscellaneous Publication 1339. United States Department of Agriculture, Forest
Service, Washington, D.C.
Hindmarch, T. D., and M. L. Reid. 2001. Thinning of mature lodgepole pine stands increases
scolytid bark beetle abundance and diversity. Canadian Journal of Forest Research,
31: 1502-1512.
Hopping, G. R., and W. G. Mathers. 1945. Observations on outbreaks and control of the
mountain pine beetle in the lodgepole pine stands of Western Canada. The Forestry
Chronicle, 21:98-108.
Ihaka, I., and R. Gentleman. 1996. R: a language for data analysis and graphics. Journal of
Computational and Graphical Statistics, 5: 299-314.
Jones, C. G., J. H. Lawton, and M. Shachak. 1997. Positive and negative effects of organisms
as physical ecosystem engineers. Ecology, 78: 1946-1957.

101

Kelsey, R. G., and G. Joseph. 2001. Attraction of Scolytus unispinosus bark beetles to ethanol
in water-stressed Douglas-fir branches. Forest Ecology and Management, 144:
229-238.
Larsson, S., R. Oren, R. H. Waring, and J. W. Barrett. 1983. Attacks of mountain pine beetle
as related to tree vigor of ponderosa pine. Forest Science, 29: 395-402.
Lundquist, J. E. 1995. Pest interactions and canopy gaps in ponderosa pine stands in the
Black Hills, South Dakota, USA. Forest Ecology and Management, 74: 37-48.
Mattson, W. J., and R. A. Haack. 1987. The role of drought in outbreaks of plant-eating
insects. BioScience, 37: 110-118.
McGregor, M. D., G. D. Amman, R. F. Schmitz, and R. D. Oakes. 1987. Partial cutting
lodgepole pine stands to reduce losses to the mountain pine beetle. Canadian Journal
of Forest Research, 17: 1234-1239.
Miller, D. R., and J. H. Borden. 1985. Life history and biology of Ips latidens (Leconte)
(Coleoptera: Scolytidae). The Canadian Entomologist, 117: 859-871.
Mitchell, R. G., R. H. Waring, and G. B. Pitman. 1983. Thinning lodgepole pine increases
tree vigor and resistance to mountain pine beetle. Forest Science, 29: 204-211.
Negron, J. F. and J. B. Popp. 2004. Probability of ponderosa pine infestation by mountain
pine beetle in the Colorado Front Range. Forest Ecology and Management, 191: 1727.
Oliver, C. D., and B. C. Larson. 1996. Forest stand dynamics. John Wiley and Sons Inc., New
York, NY.
Paine, T. D., M. C. Birch, and P. Svihra. 1981. Niche breadth and resource pardoning by four
sympatric species of bark beetles (Coleoptera: Scolytidae). Oecologia, 48: 1-6.
Paine, T. D., K. F. Raffa and T. C. Harrington. 1997. Interactions among scolytid bark beetles,
their associated fungi, and live host conifers. Annual Review of Entomology, 42:
179-206.
Poland, T. M. 1997. Competitive interactions between the spruce beetle Dendroctonus
rufipennis Kirby, and two secondary species Ips tridens Mannerheim and Dryocoetes
affaber Mannerheim (Coleoptera: Scolytidae). PhD Thesis. Simon Fraser University.
155p.

102

Raffa, K. R, B. H. Aukema, B. J. Bentz, A. L. Carroll, J. A. Hicke, M. G. Turner, and W. H.
Romme. 2008. Cross-scale drivers of natural disturbances prone to anthropogenic
amplification: the dynamics of bark beetle eruptions. Bioscience, 58: 501-517.
Raffa, K. F., and A. A. Berryman. 1986. A mechanistic computer model of mountain pine
beele populations interacting with lodgepole pine stands and its implications for forest
managers. Forest Science, 32: 789-805.
R Development Core Team. 2008. R: a language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria.
Rudinsky, J. A. 1962. Ecology of Scolytidae. Annual Review of Entomology, 7: 327-348.
Safranyik, L., and A. L. Carroll. 2006. The biology and epidemiology of the mountain pine
beetle in lodgepole pine forests, pp. 3-66. In L. Safranyik and B. Wilson (eds.), The
mountain pine beetle: a synthesis of its biology, management and impacts on
lodgepole pine. Pacific Forestry Centre, Canadian Forest Service, Natural Resources
Canada, Victoria, British Columbia.
Safranyik, L., T. L. Shore, A. L. Carroll, and D. A. Linton. 2004. Bark beetle (Coleoptera:
Scolytidae) diversity in spaced and unmanaged mature lodgepole pine (Pinaceae) in
southeastern British Columbia. Forest Ecology and Management, 200: 23-38.
Safranyik, L., D. M. Shrimpton, and H. S. Whitney. 1974. Management of lodgepole pine to
reduce losses from the mountain pine beetle. Forestry Technical Report 1. Pacific
Forestry Centre, Canadian Forest Service, Natural Resources Canada, Victoria,
British Columbia. 29p.
Sartwell, C , and R. E. Stevens. 1975. Mountain pine beetle in ponderosa pine prospects for
silvicultural control in second-growth stands. Journal of Forestry, 73: 136-140.
Savoie, A., J. H. Borden, H. D. Pierce Jr., R. Gries, and G. Gries. 1998. Aggregation
pheromone of Pityogenes knechteli and semiochemical-based interactions with three
other bark beetles. Journal of Chemical Ecology, 24: 321-337.
Schwerdtfeger, F. 1956. Scolytidae (Col.) an P/m/.v-Arten in Mittlcamerika: I. Das genus Ips
De Geer. Zeitschrift fur Angewandte Entomologie, 39: 34-57.
Smith G. D., A. L. Carroll, and B. S. Lindgren. 2009. Life history of a secondary bark beetle,
Pseudips mexicanus (Coleoptera: Curculionidae: Scolytinae), in lodgepole pine in
British Columbia. The Canadian Entomologist, 141: 56-69.

103

Steed, B. E., and M. R. Wagner. 2008. Seasonal pheromone response by Ipspini in northern
Arizona and western Montana, U.S.A. Agricultural and Forest Entomology, 10:
189-203.
Stoyan, D., and A. Penttinen. 2000. Recent applications of point process methods in forestry
statistics. Statistical Science, 15: 61-78.
Whitehead, R. J., and G. L. Russo. 2005. "Beetle-proofed" lodgepole pine stands in interior
British Columbia have less damage from mountain pine beetle. Information Report
BC-X-402. Natural Resources Canada, Canadian Forest Service, Pacific Forestry
Centre, Victoria, British Columbia, Canada.
Wood, S. L. 1982a. The role of pheromones, kairomones, and allomones in the host selection
and colonization behavior of bark beetles. Annual Review of Entomology, 27:
411-446.
Wood, S. L. 1982b. Editor. The bark and ambrosia beetles of North and Central America
(Coleoptera: Scolytidae), a taxonomic monograph. Brigham Young University, Provo,
Utah. 1359p.

104

CHAPTER 5
General Conclusions

This study was undertaken to examine the temporal and spatial relationships between
a bole-infesting asemblage of bark beetles, including the mountain pine beetle, Dendroctonus
ponderosae Hopkins, as the latter was transitioning from the endemic to the incipientepidemic phase. The extent of research pertaining to the endemic phase of this organism is
severely limited, and finding endemic beetles often poses a significant challenge (Tkacz and
Schmitz 1986, Bartos and Schmitz 1998, Carroll et al. 2006). Only recently were endemic
populations defined to be approximately forty beetles per hectare (Carroll et al. 2006), thus
contextualizing much existing research on low populations of mountain pine beetle as more
relevant to the incipient phase. In the endemic phase mountain pine beetle is believed to
inhabit weakened, diseased, and suppressed hosts (Blackman 1931, Furniss and Carolin
1977). Associations between low levels of mountain pine beetle and other bark beetles have
been noted anecdotally by DeLeon (1934), Ham el and McGregor (1976), Gohecn and Cobb
(1980), Tkacz and Schmitz (1986), and Amman and Scmitz (1988).
Temporal relationships between non-eruptive species of bark beetles and mountain
pine beetle in the first data chapter suggest facilitation between species. The number of trees
colonized by mountain pine beetle increased in relation to the number of trees colonized by
other species of bark beetles, trailing populations of other bark beetles by at least a year.
Spatial analyses from the second data chapter determined that endemic mountain pine beetle
frequently colonize hosts containing other bark beetle species and/or bearing some sort of

105

injury. The greatest number of positive spatial associations was with Pseudips mexicanus
(Hopkins), however, mountain pine beetle were found with a variety of bark beetle species
including Orthotomicus latidens (LeConte), Hylurgops spp., and less frequently Ips pini
(Wood). The colonization of trees utilised previously by other bole-infesting bark beetles
(Hamel and McGregor 1976, Furniss and Carolin 1977, Goheen and Cobb 1980, Amman and
Schmitz 1988, Carroll et al. 2006, Safranyik and Carroll 2006) may enable the insect to evade
a strong defensive response by the host. As the number of colonizations by mountain pine
beetle increased towards incipient-epidemic levels, the associations with other bark beetles
and injured trees became less evident. These results are consistent with a model in which the
colonization dynamics of mountain pine beetle change as populations increase, such that
mountain pine beetle subsist in these "nurse trees" until suitable conditions occur to strip- or
mass-attack healthy hosts.
Spatial analyses of the colonizations by secondary bark beetles in the final data
chapter suggest that the bole-infesting community likely partitions host resources effectively
to reduce interspecific competition. Furthermore, resource sharing between secondary bark
beetles within the bole-infesting assemblage may enable or prolong localized eruptions
following an epidemic of mountain pine beetle and contribute to the mortality of smaller, but
otherwise healthy trees. It is likely that competition between mountain pine beetle and other
bark beetles such as /. pini increases with the increasing density of mountain pine beetle in
later population phases (Rankin and Borden 1991).

106

Management Implications
In summary, the interaction between mountain pine beetle and other bark beetles may
be an important element to incorporate into future models predicting the risk of outbreaks of
mountain pine beetle. Futhermore, the populations of other bark beetles may potentially be
used to monitor for the development of growing populations of mountain pine beetle.
Pheromone traps could be used to monitor the populations of bark beetles that appear to be
very closely associated with endemic levels of mountain pine beetle such as P. mexicanus
and O. latidens, within stands of particular importance. However, potential challenges arise,
as the pheromone of P. mexicanus needs to be elucidated further (Smith et al. 2009).
Alternatively, the interaction between mountain pine beetle and other bark beetles
may be manipulated in such a way as to reduce the ability for stands to harbour endemic
levels of mountain pine beetle, possibly through thinning operations that reduce the number
of weakened and injured trees within a stand. However, these operations must take into
consideration the potential for immigration of populations of mountain pine beetle from
neighbouring areas, as the insect is capable of aeolian dispersal over vast distances (de la
Giroday et al. 2010). It is suspected that high levels of immigrating beetles can overwhelm
local dynamics within a stand, as likely occurred in stand E.
Future Research
Future work exploring the density-dependent relationship between secondary bark
beetles and endemic level mountain pine beetle may shed considerable light on the question
of facilitation. A further look at the nutritional changes of phloem tissue as well as the

107

moisture content of host trees innoculated by secondary bark beetle fungi may also provide
significant understanding of facilitative relationships. Future work should also focus on the
nature of interactions between species of non-eruptive bark beetles. Competition experiments
between varying densities of secondary bark beetles may lend valuable inference concerning
their relationships. A greater understanding of the extent of resource partitioning or
competition at a tree level may enhance the projection of mortality to small trees following
outbreaks of mountain pine beetle. The existence of stands in which bark beetles have
thinned trees to a level such that they are essentially immune to localized eruptions suggests
that incorporating natural processes in approaches to future management has promise.
Finally, the potential interactions of mountain pine beetle with bark beetle species found in
the boreal forest may be of interest as the threat of populations spreading into new habitats is
explored.

108

LITERATURE CITED
Amman G. D., and R. F. Schmitz. 1988. Mountain pine beetle: lodgepole pine interactions
and strategies for reducing tree losses. Ambio, 17: 62-68.
Bartos, D. L., and R. F. Schmitz. 1998. Characteristics of endemic level mountain pine beetle
populations in south central Wyoming. Research Paper RMRS-RP-13. United States
Department of Agriculture, Forest Service, Rocky Mountain Research Station, Ogden,
UT.
Blackman, M. W. 1931. The Black Hills beetle (Dendroctonus ponderosae Hopkins). N.Y.
State College of Forestry, Syracuse, N.Y. Technical Publication No. 36.
Carroll, A. L., B. H. Aukema, K. F. Raffa, G. D. Smith, and B. S. Lindgren. 2006. Mountain
pine beetle outbreak development: the endemic - incipient transition. Mountain Pine
Beetle Initiative, Project 1.03 Working Paper, Natural Resources Canada, Canadian
Forest Service. 21p.
de la Giroday, H.-M., A. L. Carroll, and B. H. Aukema. 2010. Breach of the northern Rocky
Mountain geoclimatic barrier: Initiation of range expansion by the mountain pine
beetle. Presubmission review.
DeLeon, D. 1934. An annotated list of the parasites, predators, and other associated fauna of
the mountain pine beetle in western white pine and lodgepole pine. The Canadian
Entomologist, 66: 51-61.
Furniss, R. L., and V. M. Carolin. 1977. Western Forest Insects. United States Forest Service
Miscellaneous Publication 1339. United States Department of Agriculture, Forest
Service, Washington, D.C.
Goheen, D. J., and F. W. Cobb Jr. 1980. Infestation of ceratocystis wageneri-infected
ponderosa pines by bark beetles (Coleoptera: Scolytidae) in the central Sierra
Nevada. The Canadian Entomologist, 112: 725-730.
Hamel, D. R., and M. D. McGregor. 1976. Biological notes on the emergence of mountain
pine beetle and associates from lodgepole pine, Gallatin ranger district, Gallatin
National Forest, Montana 1975. Report 76-7. United States Department of Agriculture,
Forest Service, Northern Region, Missoula, MT.
Rankin, L. J., and J. H. Borden. 1991. Competitive interactions between the mountain pine
beetle and the pine engraver in lodgepole pine. Canadian Journal of Forest Research,
21: 1029-1036.
109

Safranyik, L., and A. L. Carroll. 2006. The biology and epidemiology of the mountain pine
beetle in lodgepole pine forests, pp. 3-66. In L. Safranyik and B. Wilson (eds.), The
mountain pine beetle: a synthesis of its biology, management and impacts on
lodgepole pine. Pacific Forestry Centre, Canadian Forest Service, Natural Resources
Canada, Victoria, British Columbia.
Smith G. D., A. L. Carroll, and B S. Lindgren. 2009. Life history of a secondary bark
beetle, Pseudips mexicanus (Coleoptera: Curculionidae: Scolytinae), in lodgepole
pine in British Columbia. The Canadian Entomologist, 141: 56-69.
Tkacz, B. M., and Schmitz, R. F. 1986. Association of an endemic mountain pine beetle
population with lodgepole pine infected by armillaria root disease in Utah. Research
Note INT-353. United States Department of Agriculture, Forest Service,
Intermountain Research Station, Ogden, UT.

110

APPENDICES

APPENDIX A
Table A. 1: Mensurational characteristics of seven lodgepole pine stands at two sites in
southern British Columbia used to assess the dynamics of mountain pine beetle and
secondary bark beetle populations.
Site
Stand Area
Density
Lodgepole pine3 Secondary Trees Attacked
Angstad

Aberdeen

(Ha)

(Stems/Ha)

(Percent)

Attacks

(Percent)"

A

16.7

1263

92.5

922

1.2-4.7

B

17.6

1273

94.2

490

0.6 - 2.3

C

9.9

1554

96.8

129

0.2 - 0.9

F

13.0

1325

90.4

393

0.6 - 2.5

D

11.9

807

80.1

945

3.1 - 12.3

E

17.9

1257

90.6

434

0.5-2.1

G

14.9

1424

94.2

669

0.8-3.3

'' Remaining species included Douglas fir {Pseudotsuga menziesii), Ponderosa pine {Pinus
ponderosa), interior spruce (Picea spp.) and subalpine fir (Abies lasiocarpa)
b
First number represents a conservative estimate, second number a high estimate that does
not take into account multiple attacks on a single tree. Number represents only trees attacked
by species other than mountain pine beetle.

111

APPENDIX B
Table B.l: Number of trees colonized by year for species of the bole-infesting bark beetle
assemblage in a lodgepole pine stand of southern British Columbia (Stand A).
1999 2000 2001 2002 2003 2004
Species
2005 Total
Dendroctonus ponderosae
5
16
37
29
117
184
383
771
Ips pini

4

1

12

21

14

1

3

56

Hylurgops spp.

0

6

43

27

20

11

4

101

Orthotomicus latidens

1

2

21

76

58

32

38

228

Pseudips mexicanus

4

11

79

128

95

79

47

444

Table B.2: Number of trees colonized by year for species of the bole-infesting bark beetle
assemblage in a lodgepole pine stand of southern British Columbia (Stand B).
Species
1999 2000 2001 2002 2003 2004 2005 Total
Dendroctonus ponderosae
85
5
19
41
201
2203 3100
546
Ips pini

0

1

3

11

14

0

0

29

Hylurgops spp.

0

17

43

23

20

0

0

103

Orthotomicus latidens

0

0

15

58

33

0

0

106

Pseudips mexicanus

0

9

59

96

62

0

1

227

Table B.3: Number of trees colonized by year for species of the bole-infesting bark beetle
assemblage in a lodgepole pine stand of southern British Columbia (Stand C).
Species
2000
2001
2003
2002
2004
2005
Total
Dendroctonus ponderosae

8

10

9

36

86

270

419

Ips pini

0

1

2

0

0

0

3

Hylurgops spp.

6

12

0

0

0

1

19

Orthtomicus latidens

0

3

13

2

4

2

24

Pseudips mexicanus

5

24

20

10

10

7

76

112

Table B.4: Number of trees colonized by year for species of the bole-infesting bark beetle
assemblage in a lodgepole pine stand of southern British Columbia (Stand D).
Species
2000
2001
2002
2003
2004
2005
Total
Dendroctonus ponderosae

0

5

12

28

37

11

93

Dendroctonus murrayanae

0

10

11

19

26

24

90

Ips pint

0

0

1

6

1

2

10

Hylurgops spp.

0

6

24

22

8

4

64

Orthomicus latidens

1

6

38

126

60

52

283

Pseudips mexicanus

3

21

82

180

11

82

479

Table B.5: Number of trees colonized by year for species of the bole-infesting bark beetle
assemblage in a lodgepole pine stand of southern British Columbia (Stand E).
Total
Species
2000
2001
2002
2003
Dendroctonus ponderosae

0

19

59

404

482

Ips pini

0

2

3

13

18

Hylurgops spp.

0

11

32

61

104

Orthotomicus latidens

1

5

26

71

103

Pseudips mexicanus

1

11

36

96

144

Table B.6: Number of trees colonized by year for species of the bole-infesting bark beetle
assemblage in a lodgepole pine stand of southern British Columbia (Stand F).
Species
2001
2002
2003
2004
2005
Total
Dendroctonus ponderosae

2

7

46

84

232

371

Ips pini

0

0

1

4

8

13

Hylurgops spp.

0

2

4

12

2

20

Orthotomicus latidens

0

11

19

45

38

113

Pseudips mexican us

0

15

37

111

46

209

113

Table B.7: Number of trees colonized by year for species of the bole-infesting bark beetle
assemblage in a lodgepole pine stand of southern British Columbia (Stand G).
Species

2000

2001

2002

2003

2004

2005

Total

Dendroctoniis ponderosae

0

0

8

27

50

7

92

Dendroctoniis miirrayanae

0

0

0

2

9

12

23

Ips pini

0

0

0

1

6

4

11

Hylurgops spp.

1

1

2

14

20

12

50

Orthomicus latidens

0

0

4

53

127

58

242

Pseudips mexicanus

1

2

15

75

158

76

327

114

APPENDIX C
Table C.l: Association of trees colonized by other bark beetles on the locations of trees stripattacked by mountain pine beetle from 2001 to 2005 in a lodgepole pine stand of southern
British Columbia (Stand A). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
2
f-value
Insect
Year
Intercept
Slope
AIC

x

Strip attack
All secondaries
H. spp.
O. latidens

2000
2000
2000
2000

Est.
-9.99
-10.81
-10.57
-10.31

SE
0.33
0.54
0.48
0.40

Est.

SE

Strip attack
0. latidens
I. pini

2001
2000
2001

-9.35 0.24
-9.64 0.29
-8.82 0.27

Strip attack
P. mexicanus
All secondaries
0. latidens
H. spp.

2002
2002
2002
2002
2002

-8.96
-10.88
-10.44
-9.84
-9.70

0.02
0.61
0.54
0.44
0.41

2401 616
952 282
1885 735
6258 2606

Strip attack
P. mexicanus
I. pini

2003
2003
2003

-8.57 0.16
-9.66 0.61
-8.93 0.25

2038 1038
4340 1997

Strip attack
P. mexicanus
P. mexicanus

2004
2003
2004

-8.72 0.18
-7.64 0.45
-9.41 0.41

Strip attack
/. pini
H. spp.

2005
2005
2004

-7.94 0.12
-7.79 0.13
-7.64 0.17

5228 1906
12121 4749
14927 5747

13831 4344
-2583 3386

-2296
1508

957
745

-12255 6297
-5651 2593

115

0.01
0.02
0.03

199.76
195.29
196.67
197.00

0.01
0.05

353.90
348.70
352.03

15.44
10.99
6.19
5.59

<0.0001
0.001
0.01
0.02

500.25
486.82
491.26
496.06
496.66

4.16
4.13

0.04
0.04

710.39
708.24
708.27

0.02
0.05

623.96
620.33
622.14

0.02
0.02

1252.94
1249.93
1249.51

6.47
5.10
4.77

7.22
0.62

5.64
3.83

5.03
5.45

Table C.2: Association of trees colonized by other bark beetles on the locations of trees stripattacked by mountain pine beetle from 2001 to 2005 in a lodgepole pine stand of southern
British Columbia (Stand C). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
Insect

Year

Strip attack
H. spp.
H. spp.

2001
2001
2000

Intercept
Est. SE
-9.77 0.45
-12.85 1.44
-10.80 0.75

Slope
Est. SE

Strip attack
H. spp.
All secondaries
P. mexicanus

2002
2001
2001
2001

-9.77 0.45
-12.19 1.30
-12.20 1.29
-10.74 0.75

12562 4838
12562 4838
2642 1171

Strip attack
All secondaries
P. mexicanus
P. mexicanus
All secondaries
0. latidens
0. latidens

2003
2003
2003
2002
2002
2003
2002

-9.18
-11.18
-11.12
-10.34
-10.52
-10.23
-10.55

0.33
0.83
0.82
0.57
0.61
0.58
0.70

6847 1695
8022 2020
2979 731
2190 539
21915 6201
7055 2396

Strip attack
P. mexicanus
All secondaries
All secondaries
P. mexicanus
0. latidens
Strip attack
0. latidens
All secondaries
P. mexicanus
All secondaries
P. mexicanus
(). latidens

2004
2004
2004
2003
2003
2004
2005
2005
2005
2004
2004
2005
2004

-8.33
-8.76
-8.69
-8.75
-8.73
-8.55
-7.20
-8.02
-7.76
-7.71
-7.66
-7.65
-7.54

0.22
0.30
0.29
0.32
0.32
0.26
0.12
0.20
0.18
0.17
0.17
0.17
0.16

15059 5035
12171 4595

x2

P-value

AIC

0.001
0.02

109.67
101.44
106.19

7.37
7.36
3.94

0.01
0.01
0.05

109.67
104.26
104.26
107.69

19.38
18.82
13.24
13.22
11.48
7.99

<0.0001
<0.0001
0.0003
0.0003
0.001
0.005

185.23
167.78
168.33
173.91
173.94
175.67
179.16

10.19
5.44

2663 873
1632 584
2267 1022
2612 1218
3722 1714

7.59
6.29
4.45
4.15
3.72

0.006
0.01
0.04
0.04
0.05

16618 1973
3102 469
2974 485
1923 318
3456 591
4953 873

64.19
37.31
30.64
29.16
28.30
24.16

<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001

JI6

393.95
388.18
389.48
391.32
391.62
392.05
1068.30
1005.53
1032.41
1039.08
1040.56
1041.42
1045.56

Table C.3: Association of trees colonized by other bark beetles on the locations of trees stripattacked by mountain pine beetle from 2003 and 2005 in a lodgepole pine stand of southern
British Columbia (Stand D). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
Insect

Year

Strip attack
P. mexicanus
All secondaries
O. latidens

2003
2003
2003
2003

Intercept
Est. SE
-8.99 0.27
-12.42 1.45
-12.30 1.45
-12.16 1.41

Strip attack
2005
D. murrayanae 2005

-10.53 0.58
-14.20 2.23

Slope
Est. SE
1818 693
933 372
2459 1002

13428 6374

117

2

X

8.61
7.28
6.70

5.19

P-value

AIC

0.003
0.01
0.01

281.80
275.02
276.34
276.92

0.02

71.20
67.97

Table C.4: Association of trees colonized by other bark beetles on the locations of trees stripattacked by mountain pine beetle from 2002 and 2003 in a lodgepole pine stand of southern
British Columbia (Stand E). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
2
Insect
Year
Intercept
Slope
P-value
AIC

x

Strip attack
H. spp.
P. mexicanus

2002
2002
2002

Strip attack
2003
H. spp.
2002
All secondaries 2003
P. mexicanus
2003
H. spp.
2003
2002
/. pini
0. latidens
2003
All secondaries 2002
0. latidens
2002
I. pini
2003

Est.

SE

Est. SE
-9.23 0.25
-11.27 0.77
-8.49 0.41

18801 5723
-4032 2201

10.95
4.52

0.001
0.03

329.30
320.37
326.81

-7.89
-11.28
-12.28
-10.24
-9.12
-8.65
-9.04
-8.79
-8.44
-8.39

13677 2521
2515 470
3495 714
2728 703
32913 8385
2292 610
1324 406
2949 1010
5217 1929

44.94
44.05
25.90
16.50
15.24
14.23
10.08
8.01
7.67

<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.0001
0.002
0.005
0.006

1086.55
1043.70
1044.59
1062.74
1072.14
1073.40
1074.41
1078.57
1080.63
1080.97

0.13
0.70
0.90
0.54
0.37
0.26
0.36
0.33
0.25
0.25

118

Table C.5: Association of trees colonized by other bark beetles on the locations of trees stripattacked by mountain pine beetle from 2003 to 2005 in a lodgepole pine stand of southern
British Columbia (Stand F). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
1
Insect
Year
Intercept
Slope
P-value
AIC

x

Est. SE
Strip Attack
H. spp.

2002
2002

-11.25 0.71
-14.29 2.67

Strip attack
All secondaries
P. mexicanus
P. mexicanus
0. latidens
All secondaries
0. latidens

2003
2003
2003
2002
2003
2002
2002

-9.05
-11.66
-11.41
-10.75
-10.43
-10.43
-9.74

0.24
0.90
0.91
0.64
0.55
0.60
0.44

Strip attack
O. latidens
P. mexicanus
All secondaries
P. mexicanus
I. pini
All secondaries

2005
2005
2004
2005
2005
2004
2004

-8.23
-9.14
-9.20
-8.90
-8.90
-8.50
-9.02

0.16
0.39
0.46
0.37
0.37
0.21
0.45

Est.

SE

97282 57658

4883
7175
13064
8301
5809
7845

1379
2270
3815
2442
1960
3570

3158 1128
1207 499
967 437
1992 931
9251 4116
637 321

119

5.26

16.78
14.83
12.87
11.11
9.74
4.67

8.06
6.08
4.87
4.59
4.14
4.08

0.02

50.99
47.74

<0.0001
0.0001
0.0003
0.001
0.002
0.03

363.78
349.11
351.07
353.03
354.78
356.16
361.23

0.01
0.01
0.03
0.03
0.04
0.04

758.55
752.76
754.74
755.95
756.23
756.68
756.74

Table C.6: Association of trees colonized by other bark beetles on the locations of trees stripattacked by mountain pine beetle from 2003 and 2004 in a lodgepole pine stand of southern
British Columbia (Stand G). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
Insect

Year

Strip attack
2003
2002
0. latidens
All secondaries 2002
2003
P. mexicanus

Intercept
Est. SE
-9.21 0.29
-10.19 0.52
-10.95 0.79
-11.52 1.07

2004
Strip attack
All secondaries 2004

-8.61 0.21
-10.44 1.00

Slope
Est. SE
17229 5342
7419 2609
3138 1252

654

120

330

x2
9.83
8.62
6.70

4.15

P-value

AIC

0.002
0.003
0.01

247.09
239.30
240.50
242.42

0.04

424.67
422.58

Table C.7: Association of trees strip-attacked by mountain pine beetle on the location of mass
attacks from 2001 to 2005 in a lodgepole pine stand of southern British Columbia (Stand A).
The line in bold represents an intercept-only model; i.e., modeling a constant density of
insects across the stand. Subsequent lines reflect whether the location of each listed insect
and year provides inference on the location of the insect studied relative to this constant
density. A positive estimate for a slope reflects positive spatial association, while a negative
estimate reflects spatial inhibition at a between-tree scale. The response variable for each
equation is log(A), where A is the estimated density of trees colonized per square meter.
Significant models are listed in order of best fit for each year.
Insect

Year

Mass attack
Strip attack

2001
2001

Intercept
Est. SE
-10.39 0.41
-11.36 0.68

Mass attack
Strip attack

2003
2003

-8.01 0.12
-10.31 0.36

Mass attack
Strip attack
Strip attack

2004
2004
2003

-7.59 0.10
-9.84 0.27
-9.35 0.27

Mass attack
Strip attack
Strip attack

2005
2005
2004

-6.49 0.06
-8.05 0.13
-6.81 0.09

Slope
Est.
SE
7581

8880

7160
7109

3221
1585

121

3065

1050

551
861

179
301

2

X

4.69

70.51

201.86
67.75

280.07
25.80

/'-value

AIC

0.03

138.70
136.01

<0.0001

1173.22
1104.73

<0.0001
<0.0001

1702.56
1502.72
1636.83

<0.0001
<0.0001

4438.12
4160.11
4414.38

Table C.8: Association of trees strip-attacked by mountain pine beetle on the location of mass
attacks from 2003 to 2005 in a lodgepole pine stand of southern British Columbia (Stand C).
The line in bold represents an intercept-only model; i.e., modeling a constant density of
insects across the stand. Subsequent lines reflect whether the location of each listed insect
and year provides inference on the location of the insect studied relative to this constant
density. A positive estimate for a slope reflects positive spatial association, while a negative
estimate reflects spatial inhibition at a between-tree scale. The response variable for each
equation is log(A), where A is the estimated density of trees colonized per square meter.
Significant models are listed in order of best fit for each year.
Insect

Year

Mass attack
Strip attack
Strip attack

2003
2003
2002

Intercept
SE
Est.
-8.24 0.21
-9.54 0.39
-8.68 0.26

Mass attack
Strip attack
Strip attack

2004
2004
2003

-7.57
-8.33
-7.93

0.15
0.25
0.20

2417
2541

492
685

Mass attack
Strip attack
Strip attack

2005
2005
2004

-6.05 0.07
-7.78 0.16
-6.60 0.11

962
1892

50
247

Slope
Est.
SE
5831 832
5734 1405

122

x1

F-value

AIC

<0.0001
0.001

427.09
384.15
418.48

20.22
11.09

<0.0001
0.001

773.30
754.68
763.82

447.28
50.05

<0.0001
<0.0001

2894.01
2446.90
2844.13

44.74
10.41

Table C.9: Association of trees strip-attacked by mountain pine beetle on the location of mass
attacks from 20023 and 2003 in a lodgepole pine stand of southern British Columbia (Stand
E). The line in bold represents an intercept-only model; i.e., modeling a constant density of
insects across the stand. Subsequent lines reflect whether the location of each listed insect
and year provides inference on the location of the insect studied relative to this constant
density. A positive estimate for a slope reflects positive spatial association, while a negative
estimate reflects spatial inhibition at a between-tree scale. The response variable for each
equation is log(A), where A is the estimated density of trees colonized per square meter.
Significant models are listed in order of best fit for each year.
2
Year
Insect
Intercept
Slope
AIC
P-value

x

Mass attack
Strip attack
Strip attack

2002
2001
2002

Est.
-8.42
-9.98
-9.52

Mass attack
Strip attack
Strip attack

2003
2003
2002

-6.23
-8.56
-7.20

SE
0.17
0.52
0.42
0.06
0.23
0.13

Est.

SE

14668 3962
9115 2736

4765
8317

397
903

18.09
12.41

208.90
93.27

<0.0001
0.0004

680.03
664.00
669.68

<0.0001
<0.0001

4655.57
4449.17
4564.80

Table CIO: Association of trees strip-attacked by mountain pine beetle on the location of
mass attacks from 2004 and 2005 in a lodgepole pine stand of southern British Columbia
(Stand F). The line in bold represents an intercept-only model; i.e., modeling a constant
density of insects across the stand. Subsequent lines reflect whether the location of each
listed insect and year provides inference on the location of the insect studied relative to this
constant density. A positive estimate for a slope reflects positive spatial association, while a
negative estimate reflects spatial inhibition at a betwcen-tree scale. The response variable for
each equation is log(A), where A is the estimated density of trees colonized per square meter.
Significant models are listed in order of best fit for each year.
Insect

Year
2004
2004

Intercept
Est.
SE
-8.09 0.15
-9.61 0.46

Mass attack
Strip attack
Mass attack
Strip attack

8337 2179

2005
2004

-6.69
-8.50

5202

0.07
0.18

Slope
Est.
SE

381

P- value

AIC

13.99

0.0002

856.43
844.75

189.28

<0.0001

2938.65
2752.61

X

Figure C.l: Ripley's K function for trees strip-attacked by mountain pine beetle in 2002 for
stand A. Observed estimate is shown by the solid black line, the upper and lower limits of the
95% confidence interval are shown by the green and blue ragged dashes respectively. The
theoretical estimate for a point process displaying complete spatial randomness is shown by
the smooth red dashes. The focal distance (r) on the x-axis is represented in metres.

124

LI ppCJ

8

n

IheotetiCLxl
o h s c i " \ sd

Figure C.2: Ripley's K function for trees strip-attacked by mountain pine beetle in 2003 for
stand A. Observed estimate is shown by the solid black line, the upper and lower limits of the
95% confidence interval are shown by the green and blue ragged dashes respectively. The
theoretical estimate for a point process displaying complete spatial randomness is shown by
the smooth red dashes. The focal distance (r) on the x-axis is represented in metres.

125

lippCJ
-«~ * - ^ -

lower

~~ «~ ~~*

Iheoretieul

Figure C.3: Ripley's K function for trees mass-attacked by mountain pine beetle in 2003 for
stand A. Observed estimate is shown by the solid black line, the upper and lower limits of the
95% confidence interval are shown by the green and blue ragged dashes respectively. The
theoretical estimate for a point process displaying complete spatial randomness is shown by
the smooth red dashes. The focal distance (r) on the x-axis is represented in metres.

126

Figure C.4: Ripley's K function for trees mass-attacked by mountain pine beetle in 2004 for
stand A. Observed estimate is shown by the solid black line, the upper and lower limits of the
95% confidence interval are shown by the green and blue ragged dashes respectively. The
theoretical estimate for a point process displaying complete spatial randomness is shown by
the smooth red dashes. The focal distance (r) on the x-axis is represented in metres.

127

APPENDIX D
Table D.l: Association of trees colonized by other bark beetles on the locations of trees
colonized by O. latidens from 2002 and 2003 in a lodgepole pine stand of southern British
Columbia (Stand B). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
Insect

Year

0. latidens
P. mexicanus
P. mexicanus
Ips pini
H. spp.
H. spp.

2002
2001
2002
2001
2002
2001

O. latidens
P. mexicanus
H. spp.
Ips pini
H. spp.

2003 -8.95
2003 -11.88
2003 -11.04
2003 -10.48
2002 -9.87

Intercept
Est.
SE
-8.39 0.13
-11.08 0.63
-10.49 0.59
-8.88 0.23
-9.12 0.33
-9.42 0.23
0.17
0.81
0.65
0.55
0.45

Slope
Est.
SE
10217 2177
4773 1200
33026 10742
6987 2652
33026 10742

9576
21130
23430
8560

128

2309
5591
7110
3488

2

X

23.64
20.70
8.65
6.67
5.64

24.34
17.23
11.58
5.78

P-value

AIC

<0.000l
<0.000l
0.003
0.01
0.02

1090.97
1069.13
1072.07
1084.12
1086.10
1087.12

<0.0001
<0.001
0.001
0.02

658.81
636.35
643.47
649.11
654.91

Table D.2:, Association of trees colonized by other bark beetles on the locations of trees
colonized by O. latidens from 2001 to 2005 in a lodgepole pine stand of southern British
Columbia (Stand C). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
2
Insect
Year
Intercept
F-value
Slope
AIC

x

0. latidens
2001
P. mexicanus 2001
H. spp.
2001

Est.
-10.28
-12.48
-12.53

SE
0.58
1.35
1.63

Est.

SE

4491
11842

1508
6171

8.64
3.97

0.003
0.05

69.67
63.01
67.67

0. latidens
2002
H. spp.
2001
P. mexicanus 2001

-8.81
-10.65
-9.70

0.28
0.72
0.46

10134 2892
2485
744

12.78
8.64

0.0003
0.003

257.11
246.21
250.35

0. latidens
2003
P. mexicanus 2003

-10.68
-12.78

0.71
1.84

0.03

48.73
46.20

0. latidens
2004
P. mexicanus 2004
P. mexicanus 2003

-9.99
-12.23
-11.83

0.50
1.34
1.19

7327 2332
7767 2984

13.33
7.83

0.0003
0.005

89.92
78.55
84.06

O. latidens
2005
P. mexicanus 2005
P. mexicanus 2004

-10.68
-12.27
-12.03

0.71
1.52
1.35

7446 3566
5472 2707

4.85
3.93

0.03
0.05

48.73
45.87
46.79

8456 4435

129

4.52

Table D.3: Association of trees colonized by other bark beetles on the locations of trees
colonized by O. latidens from 2002 to 2005 in a lodgepole pine stand of southern British
Columbia (Stand D). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
Insect

Year

0. latidens
H. spp.
P. mexicanus
P. mexicanus

2002
2002
2002
2001

Intercept
Est. SE
-7.99 0.16
-9.94 0.72
-9.52 0.63
-8.57 0.34

0. latidens
P. mexicanus
H. spp.
P. mexicanus
H. spp.

2003
2003
2003
2002
2002

-6.80
-8.60
-7.74
-7.94
-7.99

0.09
0.39
0.26
0.32
0.38

1006
4144
1410
5111

197
996
365
1523

2004
0. latidens
P. mexicanus 2004
2004
H. spp.

-7.54
-9.06
-7.96

0.13
0.55
0.21

1383
9783

457
3188

O. latidens
2005
P. mexicanus 2005
2004
II. spp.

-7.68
-8.92
-8.06

0.14
0.57
0.25

Slope
SE
Est.
8118 2871
1847 690
2592 1190

1563 668
4630 2293

130

x2

P-value

AIC

0.004
0.01
0.03

685.56
678.58
679.35
682.38

28.54
17.02
15.69
11.25

<0.0001
<0.0001
<0.0001
<0.00l

1966.48
1938.40
1949.93
1951.26
1955.70

10.60
8.64

0.001
0.003

1026.50
1017.16
1019.13

0.02
0.05

904.78
900.50
902.14

8.52
7.75
4.72

5.65
4.00

Table D.4: Association of trees colonized by other bark beetles on the locations of trees
colonized by O. latidens from 2001 to 2003 in a lodgepole pine stand of southern British
Columbia (Stand E). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
1
Year
Insect
Intercept
Slope
P-value
AIC

x

Est.
0. latidens
2001
P. mexicanus 2001
H. spp.
2001

-10.39
-14.74
-9.40

SE
0.45
1.66
0.61

Est.

SE

O. latidens
H. spp
P. mexicanus
0. latidens
P. mexicanus
H. spp.

2002
2001
2002
2001
2001
2002

-8.74
-10.68
-10.03
-9.29
-9.78
-9.95

0.20
0.58
0.45
0.29
0.43
0.69

23095
4337
13334
12742
5334

5633
1085
3950
4068
2707

0. latidens
P. mexicanus
0. latidens
P. mexicanus
H. spp.

2003
2003
2002
2002
2002

-7.74
-9.96
-8.59
-8.37
-9.00

0.12
0.49
0.24
0.23
0.42

3323
4306
2398
5554

657
915
643
1652

38789 10121
-87536 60254

131

<0.0001
0.05

115.91
100.11
113.90

16.66
16.33
9.28
8.87
4.48

<0.0001
<0.0001
0.002
0.003
0.03

508.61
493.99
494.32
501.37
501.78
506.17

27.42
20.87
13.30
13.15

<0.0001
<0.0001
0.0003
0.0003

1242.78
1217.48
1224.03
1231.59
1231.74

17.81
4.02

Table D.5: Association of trees colonized by other bark beetles on the locations of trees
colonized by O. latidens from 2002 to 2005 in a lodgepole pine stand of southern British
Columbia (Stand F). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
2
Insect
Year
Intercept
Slope
P-value
AIC

x

Est.
O. latidens
2002
P. mexicanus 2002

-9.54
-11.76

SE
0.30
0.93

0. latidens
2003
P. mexicanus 2003
0. latidens
2002

-9.00
-11.26
-9.95

0.23
0.86
0.46

O. latidens
P. mexicanus
P. mexicanus
0. latidens

2004
2004
2003
2003

-8.13
-10.39
-9.89
-9.26

0.15
0.56
0.48
0.33

2578
5571
7017

549
1255
1542

0. latidens
P. mexicanus
Ips pini
P. mexicanus
0. latidens

2005
2005
2005
2004
2004

-8.30
-10.18
-9.01
-9.87
-9.27

0.16
0.50
0.29
0.53
0.40

4957 1072
9602 2624
1865 548
2678 907

Est.

SE

16265 5206

6910 2160
10273 3480

132

0.001

233.92
224.48

<0.0001
0.003

381.82
369.06
375.39

26.55
25.75
19.75

<0.0001
<0.0001
<0.0001

823.98
799.73
800.53
806.52

24.31
13.27
12.76
9.08

<0.0001
0.0003
0.0003
0.003

708.97
686.91
697.96
698.46
702.14

11.51

14.89
8.56

Table D.6: Association of trees colonized by other bark beetles on the locations of trees
colonized by O. latidens from 2002 to 2005 in a lodgepole pine stand of southern British
Columbia (Stand G). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
1
Insect
Year
Intercept
Slope
P-value
AIC

x

Est.

SE

2002
O. latidens
P. mexicanus 2002

Est.
-10.31
-12.47

SE
0.50
1.43

0. latidens
2003
P. mexicanus 2003
P. mexicanus 2002

-7.73
-9.68
-8.84

0.14
0.49
0.33

2700
7184

584
1654

0. latidens
P. mexicanus
P. mexicanus
O. latidens

2004
2004
2003
2003

-6.85
-8.45
-7.89
-7.62

0.09
0.41
0.28
0.24

1125
1511
1544

0. latidens
P. mexicanus
0. latidens
H. spp.

2005
2005
2004
2004

-7.64
-9.52
-9.18
-8.20

0.13
0.44
0.61
0.27

2503
1353
2899

0.04

92.49
90.30

22.34
18.69

<0.0001
<0.0001

927.05
906.84
910.49

270
366
413

18.13
17.05
14.45

<0.0001
<0.0001
0.0001

1996.65
1980.85
1981.93
1984.53

496
498
1111

27.04
7.78
6.84

<0.0001
0.01
0.01

1003.86
978.97
998.23
999.71

12433 6208

133

4.20

Table D.7: Association of trees colonized by other bark beetles on the locations of trees
colonized by P. mexicanus from 2001 to 2003 in a lodgepole pine stand of southern British
Columbia (Stand B). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
Insect

Year

P. mexicanus
I. pini
H. spp.

2001
2001
2001

Intercept
Est.
SE
-8.37 0.13
-8.91 0.23
-9.63 0.49

35863 10566
6688 2397

P. mexicanus
0. latidens
P. mexicanus
H. spp.
I. pini
I. pini
H. spp.

2002
2002
2001
2002
2002
2001
2001

-7.88
-10.42
-10.14
-8.86
-9.01
-8.34
-8.98

0.10
0.48
0.47
0.26
0.32
0.18
0.38

9434
8669
9040
23104
30580
5873

P. mexicanus
I. pini
H. spp.
0. latidens
H. spp.
P. mexicanus

2003
2003
2003
2003
2002
2002

-8.32
-9.88
-9.85
-9.35
-9.07
-9.21

0.13
0.40
0.42
0.31
0.32
0.43

Slope
Est.
SE

23801
16023
6491
7127
2142

134

1615
1661
2038
5783
8429
1856

5198
3799
1571
2567
943

x2

P-value

AIC

10.50
8.21

0.001
0.004

1107.73
1099.03
1101.32

42.88
28.63
18.92
16.78
12.10
10.47

<0.0001
<0.0001
<0.0001
<0.0001
0.0005
0.001

1707.69
1666.47
1680.72
1690.43
1692.57
1697.25
1698.88

<0.0001
<0.000l
<0.0001
0.01
0.02

1157.80
1137.17
1139.46
1150.27
1152.19
1153.87

22.42
20.12
17.70
7.39
5.72

Table D.8: Association of trees colonized by other bark beetles on the locations of trees
colonized by P. mexicanus from 2000 to 2005 in a lodgepole pine stand of southern British
Columbia (Stand C). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
2
Insect
Year
Intercept
Slope
P-value
AIC

x

Est.

Est.

P. mexicanus
H. spp.

2000
2000

-9.77
-11.55

SE
0.45
0.91

SE

P. mexicanus
H. spp.
O. latidens
H. spp.
P. mexicanus

2001
2001
2001
2000
2002

-8.20
-10.42
-8.80
-8.99
-8.76

0.20
0.57
0.28
0.32
0.28

11716
9934
10103
6892

2171
1908
2163
1616

P. mexicanus
P. mexicanus
O. latidens
H. spp.
O. latidens

2002
2001
2002
2001
2001

-8.38
-9.66
-9.99
-10.29
-8.88

0.22
0.41
0.49
0.59
0.30

3188
7902
10443
8895

563
1562
2340
2163

P. mexicanus 2003
2003
O. latidens
P. mexicanus 2002

-9.07
-10.08
-9.92

0.32
0.55
0.50

P. mexicanus
O. latidens
P. mexicanus
O. latidens

2004
2004
2003
2003

-9.07
-10.00
-10.68
-10.12

0.32
0.49
0.70
0.55

P. mexicanus
O. latidens
P. mexicanus
O. latidens

2005
2005
2004
2004

-9.43
-11.52
-11.24
-10.39

0.38
1.00
0.86
0.60

17106 4442

21436
2453

8842
7106
21955

29383
6498
9003

135

5885
720

1841
1809
5885

7742
1584
2202

0.0004

109.67
99.16

31.39
19.69
16.86
12.71

<0.0001
<0.0001
<0.0001
0.0004

443.53
413.93
425.63
428.46
432.61

25.39
23.56
20.87
12.30

<0.0001
<0.0001
<0.0001
0.0004

377.24
353.67
355.51
358.19
366.76

0.001
0.003

203.48
193.28
196.45

<0.0001
<0.0001
0.0003

203.48
187.64
187.67
192.63

<0.0001
<0.0001
0.0003

148.03
129.95
130.86
136.97

12.47

12.11
8.95

17.76
16.72
12.76

20.02
19.11
13.00

Table D.9: Association of trees colonized by other bark beetles on the locations of trees
colonized by P. mexicanus from 2001 to 2005 in a lodgepole pine stand of southern British
Columbia (Stand D). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
2
Insect
Year
Intercept
Slope
P-value
AIC

x

P. mexicanus
2001
P. mexicanus 2000
O. latidens
2001
D. murrayanae 2001
2002
P. mexicanus
2002
O. latidens
H. spp.
2002
P. mexicanus 2001
O. latidens
2001
D. murrayanae 2002

Est.
-8.59
-9.23
-9.12
-9.65
-7.23
-9.00
-9.06
-7.90
-7.58
-7.66

SE
0.22
0.34
0.36
0.64
0.11
0.48
0.49
0.24
0.18
0.25

2003
P. mexicanus
O. latidens
2003
P. mexicanus 2002
H. spp.
2003
H. spp.
2002
2002
O. latidens
D. murrayanae 2002

-6.44
-8.84
-7.62
-7.36
-7.48
-7.27
-6.73

0.07
0.37
0.27
0.22
0.32
0.27
0.16

P. mexicanus
O. latidens
P. mexicanus
H. spp.
O. latidens

2004
2004
2003
2004
2003

-6.92
-8.61
-8.08
-7.33
-7.80

0.10
0.36
0.38
0.17
0.41

P. mexicanus
H. spp.
O .latidens
O. latidens

2005
2004
2005
2004

-7.23
-7.61
-8.40
-8.32

0.11
0.18
0.41
0.41

Est.

SE

13670
7797
10388

3959
3489
5406

10.81
4.51
3.86

0.001
0.03
0.05

4705
7709
2951
5516
4344

1156
1892
812
1846
2070

19.27
16.60
13.21
8.20
4.24

<0.0001
<0.0001
0.0003
0.004
0.04

404.67
395.60
401.90
402.55
1350.92
1332.65
1335.32
1338.71
1343.72
1347.68

1901
1456
4056
4490
2306
3006

267
306
834
1274
699
1393

54.53
23.82
23.26
12.38
11.66
4.51

<0.0001
<0.0001
<0.000!
0.0004
0.0006
0.03

2679.99
2625.27
2655.99
2656.54
2667.42
2668.14
2675.29

<0.0001
0.0007
0.002
0.02

1760.75
1734.48
1749.84
1751.52
1756.00

0.002
0.002
0.003

1350.92
1342.15
1342.75
1343.31

2891
663
4965
731

8968
2400
1924

136

553
200
1566
318

2753
771
650

27.02
11.56
9.88
5.40

9.77
9.17
8.62

Table D.10 Association of trees colonized by other bark beetles on the locations of trees
colonized by P. mexicanus from 2001 to 2003 in a lodgepole pine stand of southern British
Columbia (Stand E). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
Insect
P. mexicanus
O. latidens

Year

Intercept
Est. SE
2001 -9.60 0.30
2001 -10.53 0.49

mexicanus
O. latidens
H. spp.
/. pini
P. mexicanus
I. pini
O. latidens

2002 -8.42
2002 -9.84
2001 -10.43
2001 -8.02
2001 -9.21
2002 -8.93
2001 -8.720

0.17
0.39
0.50
0.20
0.36
0.32
0.23

mexicanus
O. latidens
H. spp.
/. pini
O. latidens
P. mexicanus

2003
2003
2002
2002
2002
2002

0.10
0.28
0.35
0.19
0.20
0.18

-7.44
-8.37
-8.56
-7.97
-7.92
-7.75

Slope
Est.
SE
19498

5459

6597 1305
23889 4811
-38136 15209
10091 3551
23242 10901
8206 3792

1905
4962
24266
2630
1276

137

485
1386
6660
811
570

%

10.13

25.55
24.57
7.68
7.40
4.45
4.06

15.35
14.63
13.02
9.88
4.81

P-value

AIC

0.001

235.26
227.15

<0.0001
<0.0001
0.01
0.01
0.01
0.01

680.03
656.53
657.51
674.41
674.68
677.63
678.02

<0.0001
0.0001
0.0003
0.001
0.03

1621.76
1608.56
1609.28
1610.89
1614.03
1619.10

Table D.ll: Association of trees colonized by other bark beetles on the locations of trees
colonized by P. mexicanus from 2002 to 2005 in a lodgepole pine stand of southern British
Columbia (Stand F). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
2
Insect
Year
Intercept
P-value
AIC
Slope

x

P. mexicanus
0. latidens
H. spp.

Est.
2002 -9.23
2002 -10.36
2002 -9.88

SE
0.26
0.54
0.42

P. mexicanus
O. latidens
P. mexicanus
0. latidens
H. spp

2003
2003
2002
2002
2002

-8.33
-9.67
-9.79
-9.21
-8.65

0.16
0.38
0.43
0.32
0.24

P. mexicanus
P. mexicanus
0. latidens
0. latidens
I. p'mi
H. spp.
H. spp.

2004
2003
2004
2003
2004
2003
2004

-7.23
-8.26
-8.20
-7.91
-7.50
-7.47
-7.63

0.09
0.25
0.23
0.19
0.13
0.13
0.20

P. mexicanus
0. latidens
I. pini
P. mexicanus
O. latidens
H. spp.
/. pini

2005
2005
2005
2004
2004
2004
2004

-8.1!
-9.89
-8.78
-9.41
-8.96
-8.72
-8.36

0.15
0.45
0.26
0.46
0.35
0.33
0.20

Est.

SE

11687 3923
32195 12923

8115
11499
9583
18186

3483
2697
4521
8529
7652
6768

5689
9119
1568
2394
7063
8375

138

1707
2597
2481
8470

684
530
1004
2378
2476
2864

1165
2386
484
816
3121
3921

8.83
5.91

21.78
20.93
14.55
4.33

29.25
26.99
19.38
10.56
8.21
5.48

27.10
14.41
11.20
8.86
4.76
3.79

0.003
0.02

308.95
302.22
305.14

<0.0001
<0.0001
0.0001
0.04

692.34
672.80
673.65
680.03
690.25

<0.0001
<0.0001
<0.0001
0.001
0.004
0.02

1829.13
1802.60
1804.86
1812.46
1821.29
1823.64
1826.37

<0.0001
0.0001
0.001
0.003
0.03
0.05

840.23
815.43
828.12
831.33
833.67
837.77
838.74

Table D.12: Association of trees colonized by other bark beetles on the locations of trees
colonized by P. mexicanus from 2002 to 2005 in a lodgepole pine stand of southern British
Columbia (Stand G). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
2
Insect
Year
AIC
Intercept
Slope
/•-value

x

P. mexicanus
0. latidens
P. mexicanus
0. latidens
H. spp.

2003
2003
2002
2002
2003

Est.
-7.38
-8.56
-8.19
-7.67
-7.71

SE
0.12
0.33
0.26
0.16
0.21

Est.

P. mexicanus
O. latidens
P. mexicanus
O. latidens
H. spp.
H. spp.

2004
2004
2003
2003
2004
2003

-6.63
-8.21
-7.52
-7.23
-6.97
-6.87

0.08
0.37
0.25
0.21
0.16
0.14

1389
1303
1218
1795
4447

P. mexicanus
O. latidens
O. latidens
II. spp.
/. pini
H. spp

2005
2005
2004
2004
2004
2005

-7.37
-8.61
-8.51
-7.77
-7.58
-7.55

0.11
0.31
0.51
0.23
0.16
0.15

2214
1018
2134
3545
1454

2288
5459
6654
6049

139

SE
548
1392
2303
3000

18.39
15.16
7.72
3.88

<0.0001
<0.0001
0.01
0.05

1258.95
1242.75
1245.99
1253.43
1257.26

301
328
366
672
2101

22.44
15.76
11.35
7.10
4.33

<0.0001
<0.0001
0.0007
0.01
0.04

2414.52
2394.49
2401.17
2405.59
2409.83
2412.60

449
426
966
1608
623

23.71
5.91
4.87
4.64
4.62

<0.0001
0.02
0.03
0.03
0.03

1273.69
1252.18
1269.99
1271.03
1271.26
1271.27

Table D.13: Association of trees colonized by other bark beetles on the locations of trees
colonized by Hylurgops spp. from 2001 to 2002 in a lodgepole pine stand of southern British
Columbia (Stand B). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
1
Insect
Year
Slope
P-value
AIC
Intercept

x

H. spp.
2001
P. mexicanus 2001
P. mexicanus 2000

Est.
-8.78
-10.65
-9.58

SE
0.16
0.72
0.40

H. spp.
2002
P. mexicanus 2002
P. mexicanus 2001
H. spp.
2001
/. pini
2002
2002
O. latidens
I. pini
2001

-9.45
-12.29
-12.59
-11.80
-11.39
-11.82
-10.08

0.22
1.16
1.12
0.90
0.80
1.03
0.40

Est.

SE

7248
19118

2562
7853

6268 2296
11765 3790
12623 4252
37953 13522
8887 3466
40343 17799

8.28
6.39

10.66
10.57
10.20
8.76
8.12
4.65

0.004
0.01

765.20
758.78
760.67

0.001
0.001
0.001
0.003
0.004
0.03

420.10
411.37
411.46
411.83
413.26
413.91
417.37

Table D.14: Association of trees colonized by other bark beetles on the locations of trees
colonized by Hylurgops spp. from 2000 and 2001 in a lodgepole pine stand of southern
British Columbia (Stand C). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
Insect

Year

H. spp.
2000
P. mexicanus 2000

Intercept
Est.
SE
-9.77 0.45
-10.89 0.72

H. spp.
2001
H. spp.
2000
P. mexicanus 2001

-8.89
-9.67
-9.71

0.29
0.45
0.47

Slope
SE
Est.

X

F-value

AIC

10712

8.96

0.003

109.67
102.67

0.004
0.01

239.40
233.01
234.41

9970
2338

140

2971

3034
785

8.29
6.89

Table D.15: Association of trees colonized by other bark beetles on the locations of trees
colonized by Hylurgops spp. from 2001 to 2004 in a lodgepole pine stand of southern British
Columbia (Stand D). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
Insect

Year

H. spp.
0. latidens

2001
2001

Intercept
Est. SE
-10.25 0.50
-12.95 1.61

H. spp.
0. latidens

2003
2003

-9.07
-11.48

H. spp.
O. latidens
0. latidens
P. mexicanus

2004
2004
2003
2003

-10.25 0.50
-15.30 2.51
-16.09 3.20
-15.30 -3.29

0.28
1.38

Slope
Est.
SE
40341 16308

1913

7705
4313
2594

141

1002

3230
2103
1511

x2
8.60

3.95

6.47
5.08
4.20

P-value

AIC

0.003

91.96
85.31

0.05

263.74
261.63

0.01
0.02
0.04

91.96
87.45
88.83
89.71

Table D.16: Association of trees colonized by other bark beetles on the locations of trees
colonized by Hylurgops spp. from 2002 to 2003 in a lodgepole pine stand of southern British
Columbia (Stand E). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
2
P-value
AIC
Insect
Year
Slope
Intercept

x

H. spp.
P. mexicanus
I. pini
I. pini

2002
2001
2002
2001

Est.
-8.53
-9.40
-9.18
-9.01

H. spp.
I. pini
I. pini
P. mexicanus

2003
2003
2002
2003

-7.89
-9.25
-8.49
-8.70

SE
0.18
0.38
0.35
0.29
0.13
0.34
0.25
0.46

Est.

SE

10892 3740
28507 11569
27962 11710

12233
26986
1283

2346
8396
670

1.13
5.96
5.56

34.68
10.14
3.69

0.01
0.02
0.02

612.23
606.55
608.32
608.72

<0.0001
0.001
0.05

1086.55
1053.96
1078.50
1084.95

Table D. 17: Association of trees colonized by other bark beetles on the locations of trees
colonized by Hylurgops spp. from 2003 and 2004 in a lodgepole pine stand of southern
British Columbia (Stand F). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
Insect

Year

H. spp.
2002
P. mexicanus 2002

Intercept
Est.
SE
-11.25 0.71
-15.02 2.97

H. spp.
1. pini

-9.45
-10.07

2004
2004

0.29
0.41

Slope
Est.
SE
24881 15033

17087

142

5873

X

4.02

6.15

P-value

AIC

0.04

50.99
48.98

0.01

252.92
248.84

Table D.18: Association of trees colonized by other bark beetles on the locations of trees
colonized by Hylurgops spp. from 2004 and 2005 in a lodgepole pine stand of southern
British Columbia (Stand G). The line in bold represents an intercept-only model; i.e.,
modeling a constant density of insects across the stand. Subsequent lines reflect whether the
location of each listed insect and year provides inference on the location of the insect studied
relative to this constant density. A positive estimate for a slope reflects positive spatial
association, while a negative estimate reflects spatial inhibition at a between-tree scale. The
response variable for each equation is log(A), where A is the estimated density of trees
colonized per square meter. Significant models are listed in order of best fit for each year.
2
P-value
Year
Intercept
Slope
AIC
Insect

x

2004
H. spp.
2004
I. pini
D. murrayanae 2003
P. mexicanus 2003
H. spp.
2003

Est.
-9.06
-9.83
-9.54
-10.93
-9.96

SE
0.27
0.46
0.36
0.94
0.54

H. spp.
P. mexicanus
H. spp.
/. pini
0. latidens

-9.39
-12.89
-8.35
-8.71
-10.90

0.32
1.28
0.48
0.40
0.88

2005
2005
2004
2005
2005

Est.

SE

10142 3668
18892 6439
2605 1136
14832 6445

4281 1285
-12047 5938
-36068 20339
2612 1246

143

7.65
6.58
5.47
4.87

12.57
6.02
5.25
4.36

0.01
0.01
0.02
0.03

283.63
278.02
279.08
280.19
280.79

0.0004
0.01
0.02
0.04

209.89
199.35
205.90
206.67
207.56

Table D.19: Association of trees colonized by other bark beetles on the locations of trees
colonized by /. pini from 2001 and 2003 in a lodgepole pine stand of southern British
Columbia (Stand B). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
Slope
Est.
SE

X

/•-value

AIC

I. pini
2001
P. mexicanus 2001

Intercept
Est.
SE
-11.35 0.58
-20.12 4.73

29709 14064

6.91

0.01

76.10
71.18

I. pini
2003
P. mexicanus 2003

-9.81
-11.84

6915

6.30

0.01

304.65
300.30

Insect

Year

0.27
1.04

3073

Table D.20: Association of trees colonized by other bark beetles on the locations of trees
colonized by /. pini from 2002 and 2003 in a lodgepole pine stand of southern British
Columbia (Stand E). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models arc listed in order of best fit for each year.
2
Slope
Insect
Intercept
/•-value
AIC
Year

x

2002
/. pini
P. mexicanus 2001

Est.
-10.90
-13.43

SE
0.58
1.58

I. pini
2003
H. spp.
2003
P. mexicanus 2002
H. spp.
2002
O. la tide ns
2002
2002
I. pini

-9.44
-14.22
-8.08
-12.11
-8.42
-10.39

0.28
1.68
0.52
1.34
0.51
0.59

Est.

SE

26225 11556

8929 2654
-9035 4088
11037 4895
-7790 4116
40096 18285

144

4.92

20.89
9.95
7.15
5.25
4.81

0.03

73.41
70.50

<0.0001
0.002
0.01
0.02
0.03

273.33
254.46
265.40
268.20
270.10
270.54

Table D.21: Association of trees colonized by other bark beetles on the locations of trees
colonized by /. pini from 2004 and 2005 in a lodgepole pine stand of southern British
Columbia (Stand F). The line in bold represents an intercept-only model; i.e., modeling a
constant density of insects across the stand. Subsequent lines reflect whether the location of
each listed insect and year provides inference on the location of the insect studied relative to
this constant density. A positive estimate for a slope reflects positive spatial association,
while a negative estimate reflects spatial inhibition at a between-tree scale. The response
variable for each equation is log(A), where A is the estimated density of trees colonized per
square meter. Significant models are listed in order of best fit for each year.
2
P-value
AIC
Year
Intercept
Slope
Insect

x

/. pini
H. spp.
H. spp.

2004
2003
2004

Est.
-10.55
-12.21
-13.53

/. pini
0. latidens
P. mexicanus
P. mexicanus
O. latidens

2005
2005
2005
2004
2004

-9.86
-13.43
-13.11
-13.78
-12.62

SE
0.50
0.99
1.35
0.35
1.57
1.46
1.84
1.31

Est.
31800
26674

10209
7859
4183
6583

SE
9467
8180

3636
2850
1670
2491

9.69
9.05

11.81
10.71
9.72
9.25

0.002
0.003

94.43
86.77
87.40

0.001
0.001
0.002
0.002

175.77
166.01
167.10
168.09
168.57

Table D.22: Association of trees colonized by other bark beetles on the locations of trees
colonized by /. pini from 2004 in a lodgepole pine stand of southern British Columbia (Stand
G). The line in bold represents an intercept-only model; i.e., modeling a constant density of
insects across the stand. Subsequent lines reflect whether the location of each listed insect
and year provides inference on the location of the insect studied relative to this constant
density. A positive estimate for a slope reflects positive spatial association, while a negative
estimate reflects spatial inhibition at a between-tree scale. The response variable for each
equation is log(A), where A is the estimated density of trees colonized per square meter.
Significant models are listed in order of best fit for each year.
2
P-value
Insect
Year
Intercept
Slope
AIC

x

2004
/. pini
H. spp.
2003
H. spp.
2004
D. mart nyanae •2004
P. mexicanus 2003
D. rutin riyanae•2003

Est.
-9.91
-12.13
-12.66
-11.23
-13.03
-10.55

SE
0.41
1.13
1.47
0.87
1.68
0.58

Est.

SE

12281
11209
12443
4098
22986

4369
4608
5327
1867
9209

145

8.95
8.18
5.64
5.45
4.68

0.003
0.004
0.02
0.02
0.03

132.86
125.93
126.71
129.24
129.43
130.20

APPENDIX E
R statistical code outlining the use of point process models in this thesis.
1. Read in data
StandA<read.csv("/home/koopmans/Documents/EndemicMPB/StandA/naboundA.csv" ,
header=T,na.string=".")

2. Establish Boundary
b o u n d a r y A < - o w i n ( p o l y = l i s t ( x = c ( - 3 , 770, 770, 6 0 5 , 6 0 5 , 500, 500, 3 1 5 , 3 1 5 ,
155, 1 5 5 , 5 5 , 5 5 , - 3 ) , y = c ( 5 0 , 50, 190, 190, 3 2 5 , 3 2 5 , 300, 300, 350, 350,

390,

390, 315, 315))) # each boundary was different and was derived at plot set up

3. Subset by year and species
A99<-standA[standA$Att.Year=="i999", ] # Subset all attacks from 1999
mpbA9 9<-A9 9[A9 9$Att.spp=="MPB", ] # mountain pine beetle (1999)
ilatA99<-A99 [A99$Att. spp=="iLAT", ] # Orthotomicus latidens (1999)
imexA9 9<-A9 9[A9 9$Att.spp=="lMEX", ] # Pseudips mexiccinus (1999)
ipinA99<-A99[A99$Att.spp=="IPIN",] # Ips pint (1999)
dmurA99<-A99[A99$Att.spp=="DMUR, ] # D. murrayancie (1999)
hsppA9 9<-A99[A9 9 $Att.spp=="HPOR"|A9 9 $Att.spp=="HRUG"|A9 9 $Att.spp=="HSPP"]

#Hylurgop,s species (1999)
4. Create a point process and density surface
ilatA9 9ppp<-ppp(x=ilatA9 9 [ ! ( i s . n a ( i l a t A 9 9$absX)),"absX"],y=ilatA9 9 [ !
( i s . n a ( i l a t A 9 9$absY) ) , "absY" ] ,window-boundaryA) # excludes na values, incorporates
a density
#surface for colonization by O. latidens
ilatA99d<-density.ppp(ilatA99ppp, eps = l) # density measured in 1 x 1 m spacing
5. Create a homogenous point process of mountain pine beetle attack
mpbA9 9ppp<-ppp(x=mpbA99[!(is.na(mpbA9 9$absX)),"absX"],y=mpbA9 9 [ !
(is.na(mpbA99$absY) ) , "absY" ] ,window=boundaryA) # this incorporates all types of

#mountain pine beetle attack, strip attack and mass
#atlack were subsetted to be measured independently

146

6. Create a model of mountain pine beetle attack in 1999 incorporating the location of trees
attacked by O. latidens in 1999
ppmMPBilatA9 9<-ppm(mpbA9 9 p p p , - l + i l a t , c o v a r i a t e s = l i s t ( i l a t = i l a t A 9 9 d ) )

7. Determine the coefficient estimate, standard error, and AIC value for the homogenous
model of mountain pine beetle attack
coef(mpbA99ppm)
sqrt(diag(vcov(mpbA99ppm)))
AIC(mpbA9 9ppm)

8. Compare the homogenous model and the model incorporating O. latidens attack and
determine its coefficient estimate, standard error, and AIC value
anova(mpbA00ppm,ppraMPBilatA00,test="Chisq")
coef(ppmMPBilatAOO)
sqrt(diag(vcov(ppmMPBilatAOO)))
AIC(ppmMPBilatAOO)

147

APPENDIX F
Study of the effect of Pseudips mexicanus on host selection behaviour of Dendroctonus
ponderosae in cut bolts using no-choice bioassays in the laboratory

INTRODUCTION
Epidemic-level mountain pine beetle Dendroctonus ponderosae (Hopkins) are
believed to land at random and sample the host to determine its suitability for reproduction
(Hynum and Berryman 1980, Raffa and Berryman 1982, Pureswaran and Borden 2003).
Random landing may be a successful method of host searching for mountain pine beetle at
epidemic levels, as host defenses can be overcome by the attraction of conspecifics through
aggregation pheromones (Geiszler et al. 1980a, Wood 1982a, Raffa and Berryman 1983).
However, endemic level mountain pine beetle likely cannot rely on the help of mass attack by
conspecifics to overcome the defenses of healthy trees, potentially making random landing
energetically unfavorable.
Endemic-level mountain pine beetle may rely on secondary bark beetles for the
establishment of populations capable of mass attacking trees (Carroll et al. 2006; Chapter 3).
Host trees that also may be suitable for endemic level mountain pine beetle are potentially
stressed by drought events, windthrow, root disease, and/or fire damage (Geiszler et al.
1980b, Tkacz and Schmitz 1986).
The different physiological conditions of hosts utilized by epidemic and endemic
beetles, in concert with the sheer differences in population numbers, suggest that mountain

148

pine beetle may rely on different host-seeking behaviour at different population densities.
Mountain pine beetle at epidemic levels have been shown to use host volatiles in a synergistic
manner with aggregation pheromones to locate potential host trees (Borden et al. 1998).
Furthermore, volatiles of non-hosts have been shown to disrupt host-seeking behaviour
(Borden et al. 1998, Huber et al. 2000, Huber and Borden 2003). Therefore, endemic
mountain pine beetle may also be capable of utilizing volatiles emanating from potential
hosts to locate trees which are most suitable for reproductive success.
I propose that at the endemic level, mountain pine beetle orient toward trees which
have been previously infested with secondary bark beetles, or may be more likely to accept
hosts that have been partially colonized by another species of bark beetle. In light of recent
work indicating a close relationship between mountain pine beetle and Pseudips mexicaniis
(Hopkins) (Carroll et al. 2006), only one known assay has been conducted involving P.
mexicaniis related volatiles or pheromones as a primary attractant (Smith 2008). Smith found
little response by mountain pine beetle to pheromones of P. mexicaniis, but endemic
mountain pine beetle appear to benefit from associations with this insect (Smith 2008).
In the present study, mountain pine beetle was introduced to lodgepole pine hosts,
either colonized or uncolonized previously by P. mexicaniis, in a no-choice assay. Some
mountain pine beetle were starved to simulate endemic conditions, while other mountain pine
beetle were fed prior to lab assays to simulate epidemic conditions. My hypothesis is that
females of "endemic" mountain pine beetle will accept bolts previously colonized by P.
mexicamts more readily than well-fed "epidemic" beetles. These results may lend information

149

on the host acceptance behaviour of endemic level mountain pine beetle.
METHODS
Lodgepole pine trees used for host material were harvested from Telkwa, British
Columbia. Trees lacked recent exposure to epidemic populations of mountain pine beetle and
were obtained from healthy pine stands free of mountain pine beetle and secondary bark
beetle attack, as well as any signs of infection. Trees were selected on the basis of similar
growing conditions, health, vigour, and had a diameter at breast height of between 28 and 33
cm (dbh). The cut ends of all sections were sealed with hot parafin wax post-harvest to
prevent desiccation of the phloem tissue.
Mountain pine beetles used in the host selection trials were obtained from colonized
lodgepole pine trees harvested near Smithcrs, British Columbia. Trees contained teneral
mountain pine beetle adults as well as late instar larvae. The colonized pine bolts were
placed in emergence containers in a controlled environmental held at 22°C. Emerging beetles
to be used in assays were collected daily and separated based on sex (Lyon 1958). Insects
were stored at 7-8°C and were supplied with fresh pine phloem and moistened Kimwipes©
until used in experimentation, at which point beetles simulating "endemic" populations were
starved for a period of 48 hours prior to testing. "Epidemic" beetles were able to feed up
until the point of experimentation.
Healthy, uninfested lodgepole pine logs were cut into ~30cm bolts and waxed on the
ends to prevent desiccation over the course of experimentation. Treatment bolts were infested
with P. mexicanus at a density of approximately 8 attacks/m2. Pseudips mexicanus were

150

collected in Lindgren funnel traps baited with a combination of racemic ipsdienol and ipsenol
bubble cap lures (Savoie et al. 1998; Smith et al. 2009) obtained from Contech Inc., Delta,
British Columbia. Male P. mexicanus were inserted into holes drilled in the bark and beetles
were left overnight. Once all males had successfully entered and frass was visible at each
hole, two female P. mexicanus specimens were introduced to each entrance hole. Bolts were
kept at 24-25°C for ten days after female P. mexicanus introduction. Bolts were placed in test
cages approximately ten hours prior to release of test mountain pine beetles at 4pm Pacific
Standard Time.
Host selection trials were conducted in the I.K. Barber Enhanced Forestry Labatory at
the University of Northern British Columbia, Prince George, British Columbia, in a regulated
environment free from competing volatiles, and inaccessible to external specimens. Mesh
cages two by two by four feet were used as experimental arenas. The experiment was
conducted in a 2 x 2 factorial design with endemic and epidemic populations of beetles, and
treated and control bolts, with treated bolts containing colonization by P. mexicanus. There
were seven replicates for each treatment, for a total of twenty-eight bolts. Ten starved
(endemic) or unslarved (epidemic) female mountain pine beetles were introduced to the
center of each cage, approximately 55 cm from experimental bolts at the end of the enclosure.
Female mountain pine beetles were left overnight and cages checked thoroughly the
following morning. Experimental bolts were subsequently placed in sealed containers until
further examination. Examination of bolts included a summary of beetles in the container,
any apparent entrance and exit holes on the bark of each bolt, and finally a thorough

151

examination of the galleries under the bark of each bolt.
RESULTS AND DISCUSSION
No significant differences could be determined between starved or unstarved beetles
and whether or not they chose to enter a host with the number of replicates conducted in this
study. However, a trend in the data was noted. Although there appeared to be no difference
in the number of starved beetles beetween control and treated hosts, unstarved beetles
appeared to enter hosts with P. mexicanus less often than control hosts (Table E.l). This may
indicate that starved or endemic level beetle may be less discriminatory in the hosts they
colonize, compared to well-fed or epidemic beetles. Further replication may yield significant
results suggesting that beetles with higher lipid content are more likely to avoid or reject
hosts with secondary colonization in favour of hosts without other species present.

Table F.l: Number of female mountain pine beetle, found in uninfested (control) or infested
(P. mexicanus) lodgepoie pine bolts (n=lO insects introduced). Starved mountain pine beetle
were used to simulate endemic conditions, and unstarved beetles were used to simulate
epidemic conditions. There were seven replications of each treatment.
Treatment type
Starved D. ponderosae (SE)
Unstarved D. ponderosae (SE)
Control
7.14 (0.34)
6.14(0.86)
P. mexicanus
6.42 (0.72)
4.42 (0.95)

152

LITERATURE CITED
Borden, J. H., I. M. Wilson, R. Gries, L. J. Chong, H. D. Pierce Jr., and G. Gries. 1998.
Volatiles from the bark of trembling aspen, Populus tremuloides Michx. (Salicaceae)
disrupt secondary attraction by the mountain pine beetle, Dendroctonus ponderosae
Hopkins (Coleoptera: Scolytidae). Chemoecology, 8: 69-75.
Carroll, A. L., B. H. Aukema, K. F. Raffa, G. D. Smith, and B. S. Lindgren. 2006.
Mountain pine beetle outbreak development: the endemic - incipient transition.
Mountain Pine Beetle Initiative, Project 1.03 Working Paper, Natural Resources
Canada, Canadian Forest Service. 21p.
Geiszler D.R., V. F. Gallucci, and R. I. Gara. 1980. Modeling the Dynamics of Mountain Pine
Beetle Aggregation in a Lodgepole Pine Stand. Oecologia, 46: 244-253.
Huber, D. P. W., and J. H. Borden. 2003. Comparative behavioural responses of Dryocoetes
confusus Swaine, Dendroctonus rufipennis (Kirby), and Dendroctonus ponderosae
Hopkins (Coleoptera: Scolytidae) to angiosperm tree bark volatiles. Environmental
Entomology, 32: 742-751.
Huber, D. P. W., R. Gries, J. H. Borden, and H. D. Pierce Jr. 2000. A survey of antennal
responses by five species of coniferophagous bark beetles (Coleoptera: Scolytidae) to
bark volatiles of six species of angiosperm trees. Chemoecology, 10: 103-113.
Hynum, B. G., and A. A Berryman. 1980. Dendroctonus ponderosae (Coleoptera:
Scolytidae): pre-aggregation landing and gallery initiation on lodgepole pine. The
Canadian Entomologist, 112: 185-191.
Lyon, R. L. 1958. A useful secondary sex character in Dendroctonus bark beetles. The
Canadian Entomologist, 90: 582-584.
Pureswaran, D. S., and J. H. Borden. 2003. Test of semiochemical mediated host specificity
in four species of tree killing bark beetles (Coleoptera: Scolytidae). Environmental
Entomology, 32: 963-969.
Raffa, K. F., and A. A. Berryman. 1982. Gustatory cues in the orientation of Dendroctonus
ponderosae (Coleoptera: Scolytidae) to host trees. The Canadian Entomologist, 114:
97-104.
Raffa, K. F., and A. A. Berryman. 1983. The role of host plant resistance in the colonization
behavior and ecology of bark beetles (Coleoptera: Scolytidae). Ecological
Monographs, 53: 27-49.
153

Savoie A., J. H. Borden, H. D. Pierce Jr., R. Gries, and G. Gries. 1998. Aggregation
pheromone of Pityogenes knechteli and semiochemical-based interactions with three
other bark beetles. Journal of Chemical Ecology, 24: 321-337.
Smith G. D., A. L. Carroll, and B S. Lindgren. 2009. Life history of a secondary bark
beetle, Pseudips mexicanus (Coleoptera: Curculionidae: Scolytinae), in lodgepole
pine in British Columbia. The Canadian Entomologist, 141: 56-69.
Tkacz, B. M., and R. F. Schmitz. 1986. Association of an endemic mountain pine beetle
population with lodgepole pine infected by armillaria root disease in Utah. Res. Note
INT-353. United States Department of Agriculture, Forest Service, Intermountain
Research Station., Ogden, UT, US, 8p.
Wood, S. L. 1982. The role of pheromones, kairomones, and allomones in the host selection
and colonization behavior of bark beetles. Annual Review of Entomology, 27:
411-446.

154