Growth Responses Of Three Coexisting Conifer Species To Climate Variables
Across A Range Of Climate Conditions

Yumiko Miyamoto
B.Sc. Thompson Rivers University, 2005

Thesis Submitted In Partial Fulfillment Of
The Requirements For The Degree Of
Master of Science
In
Natural Resources and Environmental Studies
(Forestry)

The University of Northern British Columbia
April 2008
© Yumiko Miyamoto, 2008

1*1

Library and
Archives Canada

Bibliotheque et
Archives Canada

Published Heritage
Branch

Direction du
Patrimoine de I'edition

395 Wellington Street
Ottawa ON K1A0N4
Canada

395, rue Wellington
Ottawa ON K1A0N4
Canada

Your file Votre reference
ISBN: 978-0-494-48821-8
Our file Notre reference
ISBN: 978-0-494-48821-8

NOTICE:
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.

AVIS:
L'auteur a accorde une licence non exclusive
permettant a la Bibliotheque et Archives
Canada de reproduire, publier, archiver,
sauvegarder, conserver, transmettre au public
par telecommunication ou par Plntemet, prefer,
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.

Canada

ABSTRACT
Tree-ring analyses and an interpolated climate model (ClimateBC) were used to compare
radial growth responses to climate variables among three coexisting, ecologically distinct
conifer species, including interior spruce (Picea glauca x Picea Engelmannii), lodgepole
pine (Pinus contorta var. latifolia) and subalpine fir (Abies lasiocarpa) across a range of
climate conditions in western Canada, and altitudinal treelines in the Engelmann SpruceSubalpine Fir forests in central British Columbia (BC). Ring-width chronologies were
developed and correlated with site-specific climate data in the past 50 years from 19532002. Spruce ring-width chronologies were mainly correlated with June-July
temperatures across the sample sites, while pine and fir chronologies from in BC were
correlated with October-March temperatures. Given the species- and site-specific growth
responses to climate, future climate change will likely alter interactions among coexisting
tree species, potentially leading to changes in species dominance, compositions and
distributions of forest communities.

TABLE OF CONTENTS
ABSTRACT

n

TABLE OF CONTENTS

iii

LIST OF TABLES

v

LIST OF FIGURES

vi

ACKNOWLEDGEMENTS

vii

Chapter 1. Introduction
1.1. Tree responses to climate change
1.2. Management perspectives
1.3. Study objectives
1.4. Literature cited

1
3
4
6

Chapter 2. Growth responses of three coexisting conifer species to climate variables
across wide geographic and climate ranges
Abstract
2.1. Introduction
2.2. Methods
2.2.1. Study sites
2.2.2. Chronology development
2.2.3. Climate data
2.2.4. Growth-climate relationships
2.2.5. Gradient analyses
2.3. Results
2.3.1. Chronology statistics
2.3.2. Growth-climate relationships
2.3.3. Gradient analyses
2.3.3.1. Growing season temperatures
2.3.3.2. October-March temperatures
2.4. Discussion
2.4.1. Spatial patterns
2.4.2. Gradient analyses
2.4.2.1. Growing season temperatures
2.4.2.2. October-March temperatures
2.4.3. Growth responses of ecologically distinct species
2.4.4. Implications
2.5. Literature cited

9
10
12
12
14
16
17
18
18
18
19
21
21
22
23
23
25
25
27
29
30
32

Appendix A
Appendix B
Appendix C

Mean annual temperature and mean annual precipitation of each sample
site for the period 1961-1990.
Summary characteristics of the Arstan chronology of each population.
Pearson correlation significance of ring-width indices against mean
monthly temperatures (1) and precipitation (2) from May of the
previous growth year to September of the current growth year for the
three species.

46
47

49
in

Chapter 3. Growth responses of three coexisting conifer species at altitudinal treelines
in the central interior of British Columbia.
Abstract
3.1. Introduction
3.2. Methods
3.2.1. Study sites
3.2.2. Chronology development
3.2.3. Data analyses
3.2.3.1. Principal component analysis
3.2.3.2. Climate model
3.2.3.3. Growth-climate relationships
3.3. Results
3.3.1. Chronology statistics
3.3.2. Ring-width variation explained by temperature versus
precipitation
3.3.3. Similarity among ring-width indices
3.3.4. Growth-climate relationships
3.3.4.1. Identification of key climate variables
3.3.4.2. Regression and separate slopes analyses
3.4. Discussion
3.4.1. Importance of temperature versus precipitation
3.4.2. Similarity among ring-width indices
3.4.3. Important growth-climate relationships
3.4.4. Interspecific differences and adaptive mechanisms
3.4.5. Implication and future research
3.5. Literature cited

50
51
54
54
55
57
57
58
58
60
60
60
60
61
61
62
62
63
63
64
66
67
69

IV

LIST OF TABLES
Table 2.1.

Descriptions of the sample sites.

39

Table 2.2.

Significant correlation relationships between mean monthly
temperatures and ring-width chronologies from May of the previous
growth year to September of the current growth year for the period
1953-2002.

40

Correlation relationships between monthly heat-moisture chronologies
and the ring-width chronologies of lodgepole pine populations in Yukon
from May of the previous growth year to September of the current
growth year for the period 1953-2002.

41

The coefficient of determination (R2) from the simple linear regression
analyses between ring-width chronologies and selected predictor climate
variables.

41

Table 3.1.

Locations and climate conditions of sample sites.

75

Table 3.2.

Descriptions of the Arstan chronology statistics.

75

Table 3.3.

Percent variance in ring-width chronologies explained by 17 monthly
climate variables (PRECON, version 5.11).

75

Varimax orthogonally rotated factor loadings for the
PC1-PC3 of the nine ring-width chronologies for the period 1953-2002.

76

Correlations between predictor climate variables for the 1953-2002
period.

76

Table 2.3.

Table 2.4.

Table 3.4.
Table 3.5.

v

LIST OF FIGURES
Figure 2.1.

Study locations in Yukon and British Columbia, Canada.

42

Figure 2.2.

The range of climate conditions across the sample sites (ClimateBC
version 3.2).

43

The growth sensitivities (regression coefficients: bl) of interior spruce
and lodgepole pine to the current growing season temperatures
(June-August) along summer temperature and precipitation gradients.

44

The growth sensitivities (regression coefficients; bl) of lodgepole pine
(a) and subalpine fir (b) to October- March temperatures prior to growth
along the mean annual temperature (°C) gradient.

45

Figure 3.1.

Study sites in central British Columbia, Canada.

77

Figure 3.2.

Pearson correlation coefficients between ring-width chronologies and
mean monthly temperature (a) and precipitation (b) from May of the
previous growth year to September of the current growth year
(horizontal axes) for the period 1953-2002.

78

Similarity of ring-width variability among the nine ring-width
chronologies according to the three axes resulting from a principal
component analysis with Varimax rotation.

79

Pearson correlation coefficients between ring-width indices and monthly
PDO from May of the previous growth year to September of the current
growth year (horizontal axes) for the period 1953-2002.

80

Linear regression coefficients (bl) between ring-width indices and
selected climate variables for the period 1953-2002 at the three study
sites.

81

Figure 2.3.

Figure 2.4.

Figure 3.3.

Figure 3.4.

Figure 3.5.

VI

ACKNOWLEDGEMENTS
The project was funded by the Forest Science Program (FSP Project Y072107), Ministry of
Forests British Columbia. I would like to thank financial contributions from the Canadian
Natural Sciences and Engineering Research Council and the University of Northern British
Columbia.

I would like to thank Scott Green, Kathy Lewis, Roger Wheate and Stephen Dery for advice
and comments on my thesis, Morgan Anderson, Brooke Clasby, Hardy Griesbauer, Emily
Muller and Kara Przeczek for field and lab assistance, Tongli Wang, Andreas Hamann and
Greg O'Neill for help with ClimateBC model, Robert Westfall for help with PDO data access,
and Doug Thompson and Sean Haughian for technical support.

vn

Chapter 1. Introduction
1.1. Tree responses to climate change
Instrumental records indicate that global surface air temperature increased approximately 0.6
°C in the past 150 years, with nine out of the 10 warmest years on record after 1997 (Jones et al.
1999, Brohan et al. 2006). Based on climate records, higher latitudes warmed more than lower
latitudes in the Northern Hemisphere and general circulation models project continued
warming in mean annual temperature and slight increase or little change in precipitation in the
northern continents (Jones et al. 1999). These changes could significantly affect the
distribution and growth of tree species (Woodward 1987). Changes in temperature,
precipitation and their interactions will likely alter many factors regulating tree growth,
including growing season temperature, heat-sum accumulation, the duration of growing season,
chilling requirement and soil moisture availability. Changes in humidity, evapotranspiration
and the amount of spring runoff (that is associated with winter snowpack) can also influence
moisture availability in soils.

Trees have the ability to adapt to both short- and long-term changes in climate (Linhart and
Grant 1996, Ackerly et al. 2000, Gray 2005). Many temperate and boreal tree species tend to
show a high degree of intraspecific genetic diversity due to outcrossing, wind pollination, high
fecundity and long generation times (Aitken and Hannerz 2001, Hamrick 2004). A high degree
of plasticity in trees has been demonstrated in provenance studies, whereby seeds from a single
population are planted in different environmental conditions (Rehfeldt et al. 1999). High
genetic variation and phenotypic plasticity may allow trees to respond to a greater range of
year-to-year environmental variability and extreme events over their long life spans, compared

1

with annuals or short-lived herbaceous perennials (Loehle and LeBlanc 1996, Cordell et al.
1998, Ackerly et al. 2000, Hamrick 2004).

Tree responses to climate may vary by site conditions and species, due to the considerable
variations in environmental characteristics (e.g., elevation and topography) and in adaptive
traits in trees (e.g., physiology, morphology) (Spittlehouse 1997, 2005, Walther 2003, Gray
2005). For example, high-elevation trees are thought to be more sensitive to temperature
variability while the low-elevation trees of the same species may be more sensitive to
precipitation variability (Loehle 2000). It has been suggested that the warming in the past 60
years led to the enhanced growth and upward migration of European tree species at altitudinal
treelines in the Alps (Kullman 2002), but the warming possibly induced moisture stresses on
trees at northern treelines and caused the southward recession of white spruce ranges in Alaska
and Yukon (Barber et al. 2000, D'Arrigo et al. 2004). Green (2005) reported that phenological
responses, such as advancement or delay of growth initiation and cessation, differed among
species and along elevation in northwestern conifers in Canada.

Species-specific responses to climate change may affect competitive ability, with potentially
significant impacts on forest ecosystem functioning (Hansen et al. 2001, Walther et al. 2002).
Species interactions, dominance, and compositions may change if future climate conditions
favour one species over the others. Dullinger et al. (2005), for example, reported that the
increases in abundance of Mugo pine (Pinus mugo Turra) negatively affected the recruitment
and growth of Norway spruce (Picea abies [L.] Karst.) and European larch (Larix decidua
Mill.) seedlings under warming conditions at altitudinal treelines in Austria. They suggest that
this could affect the species compositions and geographical distributions of the three
2

coexisting species in the long run.

1.2. Management perspectives
Several authors have suggested that forest managers need to develop adaptive management
strategies to minimize risks and maximize benefits that can result from climate change
(Spittlehouse 1997,2005, Stewart et al. 1998, Chuine and Beaubien 2001, Thuiller 2003,
Hamrick 2004). From wood-production perspectives, changes in wood supply may pose
economic impacts to some extent if future climate alters tree growth rate, reproductive capacity
and susceptibility to disturbances (Rehfeldt et al. 1999, Nigh et al. 2004, Wang et al. 2006). For
example, Wang et al. (2006) projected that the productivity of interior lodgepole pine in BC
will likely increase if mean annual temperature increases to certain thresholds (values are
site-specific), but will likely decrease if the warming continues beyond the thresholds. This
warming may also change wood strength by changing the proportion of thick cell-walled,
dense latewood and thin cell-walled earlywood (Denne 1976, Telewski et al. 1999).

Forest managers should ensure that planted genotypes or species are suited to the current and
future environments (Nigh et al. 2004, Spittlehouse 2005). A detailed understanding of siteand species-specific growth responses and potential adaptive patterns may help in refining
seed transfer zones and species selection guidelines for reforestation (Spittlehouse 1997,
Johnson et al. 2004). A facilitated migration, for example, is a transfer of species or genotypes
to regions where productivity may be increased under anticipated future climate conditions
(Hamann and Wang 2006, Wang et al. 2006). Wang et al. (2006), for example, projected that
the overall productivity of interior lodgepole pine increases 14-36 % if optimal seeds sources,
compared with local seed sources, are used for reforestation under a scenario in which mean
3

annual temperature increases by 2 °C in the next 50-100 years. In the absence of additional
information, however, genotype redistribution within a species range may have an advantage
over the introduction of a species to the regions outside of its current ranges, in terms of
minimizing the risk of unexpected decline in the productivity and ecosystem health (Natural
Resources Canada 2002).

1.3. Study objectives
This study was conducted to compare radial growth-climate relationships among three
coexisting, ecologically distinct northern conifer species. The study species included
shade-intolerant, early-successional interior lodgepole pine, shade-tolerant, late-successional
subalpine fir (Abies lasiocarpa [Hook.] Nutt.) and intermediate-shade-tolerant,
mid-successional interior spruce (Picea glauca (Moench) Voss x Picea Engelmannii Parry ex
Engelm.). Interior spruce included white spruce (Picea glauca [Moench] Voss), Engelmann
spruce and their hybrids (P. glauca x P. engelmannii). These three spruce are treated as the
same species in British Columbia (BC) for management purposes (Nigh et al. 2004) and we
used this practice in this study. Tree-ring analyses were used to identify important climate
variable(s) that were correlated with annual ring-widths in the past 50 years. These important
correlations (i.e., growth response) and the strength and direction of the correlations (i.e.,
growth sensitivity) were compared among and within species at different spatial scales.

Chapter two focused on populations sampled from wide geographic and climate ranges,
extending from the southern interior of BC to central Yukon. Changes in growth sensitivities
were characterized along climate gradients for each species. Species-specific growth responses
to climate were evaluated in relation to the shade-tolerance and successional positions of
4

species.

Chapter three focused on populations at altitudinal treelines, where climate changes are
expected to be large (Millar et al. 2004, Christensen et al. 2007). We examined populations at
three climatically and geographically distinct treelines in the Engelmann Spruce-Subalpine Fir
forests in central BC. We tested a hypothesis that high-elevation trees are more sensitive to
temperature than to precipitation. We also investigated potential associations between tree
radial growth and the Pacific Decadal Oscillation, a large-scale interdecadal climate variability
that is thought to influence the climate of western North America (Mantua and Hare 2002).

5

1.4. Literature cited
Ackerly, D.D., Dudley, S.A., Sultan, S.E., Schmitt, J., Coleman, J.S., Linder, C.R., Sandquist,
D.R., Geber, M.A., Evans, A.S., Dawson, T.E., and Lechowicz, M.J. 2000. The
evolution of plant ecophysiological traits: recent advances and future directions.
BioScience. 50: 979-995.
Aitken, S.N., and Hannerz, M. 2001. Genecology and gene resource management strategies for
conifer cold hardiness. In Conifer cold hardiness. Edited by F.J. Bigras and S.J.
Colombo. Kluwer Academic Publishers, Dordrecht, Netherlands, pp 23-53.
Barber, V.A., Juday, G.P., and Finney, B.R 2000. Reduced growth of Alaskan white spruce in
the twentieth century from temperature-induced drought stress. Nature. 405: 668-673.
Brohan, P., Kennedy, J.J., Haris, I., Tett S.F.B., and Jones, P.D. 2006: Uncertainty estimates in
regional and global observed temperature changes: a new dataset from 1850. J.
Geophysical Research 111, D12106, doi:10.1029/2005JD006548.
Christensen, J.H., Hewitson, B., Busuioc, A., Chen, A., Gao, X., Held, I., Jones, R., Kolli, R.K.,
Kwon, W.-T., Laprise, R., Magana Rueda, V., Mearns, L., Menendez, C.G., Raisanen,
J., Rinke, A., Sarr A., and Whetton, P. 2007. Regional Climate Projections. In: Climate
Change 2007: The Physical Science Basis. Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and
H.L. Miller (eds.). Cambridge University Press, Cambridge, United Kingdom and New
York, NY, USA.
Chuine, I., and Beaubien, E.G. 2001. Phenology is a major determinant of tree species range.
Ecology Letters. 4: 500-510.
Cordell, S., Goldstein, G, Mueller-Dombois, D., Webb, D., and Vitousek, P.M. 1998.
Physiological and morphological variation in Metrosideros polymorpha, a dominant
Hawaiian tree species, along an altitudinal gradient: the role of phenotypic plasticity.
Oecologia. 113: 188-196.
D'Arrigo, R. D., Kaufmann, R.K., Davi, N., Jacoby, G.C., Laskowski, C, Myneni., R.B., and
Cherubini, P. 2004. Thresholds for warming-induced growth decline at elevational
tree line in the Yukon Territory, Canada. Global Biogeochemical Cycles. 18: GB3021,
doi: 10.1029/2004GB002249.
Denne, M.P. 1976. Predicting differences in potential wood production from tracheid
diameters and leaf cell dimensions of conifer seedlings. In Tree physiology and yield
improvement. Edited by M.G.R. Cannell and F.T. Last. Academic Press. London, UK.
pp 281-289.

6

Dullinger, S., Dirnbock, T., Kock, R., Hochbichler, E., Englisch, T., Sauberer, N., and Grabherr,
G. 2005. Interactions among tree-line conifers: differential effects of pine on spruce and
larch. Journal of Ecology. 93: 948-957.
Gray, P.A. 2005. Impacts of climate change on diversity in forested ecosystems: some
examples. Forestry Chronicle. 81: 655-661.
Green, D.S. 2005. Adaptive strategies in seedlings of three co-occurring, ecologically distinct
northern coniferous tree species across an elevational gradient. Canadian Journal of
Forest Research. 35: 910-917.
Hamann, A., and Wang, T. 2006. Potential effects of climate change on ecosystem and tree
species distribution in British Columbia. Ecology. 87: 2773-2786.
Hamrick, J.L. 2004. Response of forest trees to global environmental changes. Forest Ecology
and Management. 197: 323-335.
Hansen, A.J., Neilson, R.P., Dale, V.H., Flather, C.H., Iverson, L.R., Currie, D.J., Shafer, S.,
Cook, R., and Bartlein, P.J. 2001. Global change in forests: responses of species,
communities, andbiomes. BioScience. 51: 765-779.
Johnson, G.R., Sorense, F.C., St Clair, J.B, and Cronn, R.C. 2004. Pacific northwest forest tree
seed zones. Native Plants. Fall: 131-140.
Jones, P.D., New, M., Parker, D.E., Martin, S., and Rigor, I.G. 1999. Surface air temperature
and its changes over the past 150 years. Reviews of Geophysics. 37: 173-199.
Kullman, L. 2002. Rapid recent range-margin rise of tree and shrub species in the Swedish
Scandes. Jouranl of Ecology. 90: 68-77.
Linhart, Y.B., and Grant, M.C. 1996. Evolutionary significance of local genetic differentiation
in plants. Annual Review of Ecoogy and Systematics 27: 237-277.
Loehle, C. 2000. Strategy space and the disturbance spectrum: a life-history model for tree
species coexistence. American Naturalist. 156: 14-33.
Loehle, C, and LeBlanc, D. 1996. Model-based assessments of climate change effects on
forests: a critical review. Ecological Modeling. 90: 1-31.
Mantua, N.J., and Hare, S.R. 2002. The Pacific Decadal Oscillation. Journal of Oceanography.
58: 35-44.
Millar, C.I., Westfall, R.D., Delany, D.L., King, J.C. and Graumlich, L.J. 2004. Response of
subalpine conifers in the Sierra Nevada, California, U.S.A., to 20th-century warming
and decadal climate variability. Arctic, Antarctic, and Alpine Research. 36: 181-200.

7

Natural Resources Canada. 2002. Climate change impacts and adaptation: a Canadian
perspective. The Climate Change Impacts and Adaptation Directorate, Natural
Resources Canada. Ontario. 20 pp.
Nigh, G.D., Ying, C.C., and Qian, H. 2004. Climate and productivity of major conifer species
in the interior of British Columbia, Canada. Forest Science. 50: 659-671.
Rehfeldt, G.E., Ying, C.C., Spittlehouse, D.L., and Hamilton Jr., D.A. 1999. Genetic responses
to climate in Pinus contorta: niche breadth, climate change, and reforestation.
Ecological Monographs. 69: 375-407.
Spittlehouse, D.L. 2005. Integrating climate change adaptation into forest management.
Forestry Chronicle. 81: 691-695.
Spittlehouse, D.L. 1997. Forest management and climate change. In: Future Climate Change in
B.C and the Yukon. E, Taylor and B. Taylor (eds.). Environment Canada, Vancouver,
B.C. pp 24-1 - 24-8.
Stewart, R.B., Wheaton, E., and Spittlehouse, D.L. 1998. Climate change: implications for the
boreal forest. In Emerging air issues for the 21 st century: the need for multidisciplinary
management. Eds. A.H. Legge and L.L. Jones. Proceedings of an International
Specialty Conference, Calgary, September 1997. pp 86-101.
Telewski, F.W., Swanson, R.T., Strain, B.R., and Burns, J.M. 1991. Wood properties and ring
width responses to long-term atmospheric C02 enrichment in field-grown loblolly
pine {Pinus taeda L.). Plant Cell and Environment. 22: 213-219.
Thuiller, W. 2003. BIOMOD- optimizing predictions of species distributions and projecting
potential future shifts under global change. Global Change Biology. 9: 1353-1362.
Walther, G-R. 2003. Plants in a warmer world. Perspectives in Plant Ecology, Evolution and
Systematics. 6: 169-185.
Walther, G-R., Post, E., Convey, P., Menzel, A., Parmesan, C, Beebee, T.J., Fromentin, J-M.,
Hoegh-Guldberg, O., and Bairlein, F. 2002. Ecological responses to recent climate
change. Nature. 416: 389-395.
Wang, T., Hamann, A., Yanchuk, A., O'Neill, G.A., and Aitken, S.N. 2006. Use of response
functions in selecting lodgepole pine populations for future climates. Global Change
Biology. 12: 2404-2416.
Woodward, A., Silsbee, D.G., Schreiner, E.G., and Means, J.E. 1994. Influence of climate on
radial growth and cone production in subalpine fir {Abies lasiocarpa) and mountain
hemlock {Tsuga mertensiana). Canadian Journal of Forest Research. 24: 1133-1143.

8

Chapter 2. Growth responses of three coexisting conifer species to climate
variables across wide geographic and climate ranges
Abstract
Tree-ring analyses were used to compare radial growth-climate relationships among three
coexisting conifer species across wide geographic and climate ranges extending from southern
British Columbia (BC) to central Yukon, Canada. Study species included interior spruce (Picea
glauca x Picea Engelmannii), lodgepole pine (Pinus contorta var. latifolia) and subalpine fir
{Abies lasiocarpa). Interior spruce was positively or negatively correlated with June-July
temperatures of the current growth year across the sample sites while lodgepole pine and
subalpine fir showed intraspecific variations in growth-climate correlations across the sites.
BC pine and fir were predominantly positively correlated with October-March temperatures
prior to growth. Our data suggest that 1) the growth-climate relationships of interior spruce
may differ distinctively from those of lodgepole pine and subalpine fir, 2) winter temperatures
prior to growth may have significant impacts on tree growth and forest communities at some
sites, and 3) the shade-tolerance and successional positions of tree species may not contribute
to predict growth-climate relationships for mature trees.

9

2.1. Introduction
Climate plays a fundamental role regulating many physiological and phenological activities of
trees (Kramer et al. 2000, Kozlowski 2002, Walther 2003). Climate change may result in
changes in the growth, survival and reproductive capacity of trees, leading to altered forest
communities, including species interaction, abundance, composition, distribution and
susceptibility to disturbances (Hanninen et al. 2001, Kramer et al. 2000, Bertrand and
Castonguay 2003). A better understanding of tree growth-climate relationships may help in
predicting the potential impacts of climate change on forest ecosystems (Cook and Cole 1991,
Hamrick 2004, Spittlehouse 2005).

Tree-ring analyses can provide long-term growth-climate relationships (Fritts 1976). A large
body of literature on tree-ring analyses suggests that radial growth responses to climate are
likely site- and species-specific (Cook and Cole 1991, Graumlich 1993, Peterson and Peterson
1994, Villalba et al. 1994, Hofgaard et al. 1999, Makinen et al. 2002, Pederson et al. 2004,
Goldblum and Rigg 2005). In western North America, for example, numerous studies reported
that the radial growth of white spruce (Picea glauca [Moench] Voss) was positively (St.
George and Luckman 2001, D'Arrigo et al. 2005) or negatively (D'Arrigo et al. 2004,
Wilmking et al. 2004) correlated with summer temperatures of the current growth year
depending on site environmental conditions of the study sites. Similarly, Splechtna et al.
(2000) reported that the radial growth of subalpine fir {Abies lasiocarpa [Hook.] Nutt.) was
negatively correlated with summer temperatures at low-elevations, but positively correlated at
high-elevations in the interior of British Columbia (BC). Species-specific growth responses to
climate among coexisting species were also reported in several forest types in North America
(Graumlich 1993, Peterson and Peterson 1994, Goldblum and Rigg 2005). For example,
10

Peterson and Peterson (1994) reported that the radial growth of Engelmann spruce was
correlated with summer temperatures but subalpine fir was correlated with winter precipitation
at high-elevation sites in the North Cascade Mountains. Coexisting tree species probably
display unique growth responses to climate because they tend to exhibit somewhat different
optima for resource requirements to coexist (May 1972, Bazzaz 1987, He et al. 2005).

Many northern conifers are reported to show clinal patterns in growth-climate relationships
along environmental gradients, such as elevation (Splechtna et al. 2000), latitude (e.g.,
Hofgaard et al. 1999, Makinen et al. 2002) and moisture condition (Linderholm 2001). For
example, studies reported that the importance of June-July temperatures on the radial growth
of Norway spruce (Picea abies [L.] Karst.) increased with elevation (Savva et al. 2006) and
with latitude (Makinen et al. 2002). Although many tree-ring studies have demonstrated
intraspecific variations in the growth-climate relationships for single species across its
distribution or multiple species within small spatial scales (Villalba et al. 1994, Peterson et al.
2002), it is not clear how these relationships change within and among species across wide
geographic and climate ranges.

Interior spruce (Picea glauca [Moench] Voss x Picea Engelmannii Parry ex Engelm.),
lodgepole pine (Pinus contorta Dougl. var. latifolia Engelm.) and subalpine fir are three
conifer species that often coexist across wide geographic and climate ranges in western North
America (Burns and Honkala 1990). Lodgepole pine is a shade-intolerant, early-successional
species that colonizes quickly after disturbance. Interior spruce is an
intermediate-shade-tolerant, mid-successional species, and subalpine fir is a shade-tolerant,
late-successional species that increases in abundance and dominance through time in the
11

absence of disturbance. Green (2005,2007) suggests that lodgepole pine and subalpine fir may
show distinct patterns in the growth-climate relationships while interior spruce may display an
intermediate pattern among the three study species because of the dissimilarities in ecological
characteristics, such as shade-tolerance and successional positions. Green (2007) reported the
potential associations between ecological characteristics and phenological responses to
seasonal weather conditions among the three coexisting species based on seedling studies in
the central interior of BC. However, a question remains if this association exists for mature
trees across wide geographic and climate ranges (Green 2007).

Specific study objectives were 1) to identify and compare climate variables most strongly
correlated with tree radial growth within and among the species of lodgepole pine, interior
spruce and subalpine fir, 2) to characterize clinal patterns in the growth sensitivities of
populations to common predictor climate variables along climate gradients for each species,
and compare the patterns among species, and 3) to evaluate these comparisons in relation to the
shade-tolerance and successional positions of species.

2.2. Methods
2.2.1. Study sites
Study sites extended from the southern interior of British Columbia (BC) to central Yukon,
Canada, covering a wide range in mean annual temperature and precipitation (Table 2.1,
Figure 2.1, 2.2, Appendix A). The sample sites in BC ranged from 650 m to an altitudinal
treeline at 1800 m in elevation and climate conditions varied considerably among the sites.
The sample sites were located within five biogeoclimatic zones, including Interior
Douglas-fir (IDF), Montane Spruce (MS), Interior Cedar-Hemlock (ICH), Engelmann
12

Spruce-Subalpine Fir (ESSF), and Sub-Boreal-Spruce (SBS) zones (Meidinger and Pojar
1991) (Table 2.1). IDF has dry and warm summers with relatively long growing seasons and
cool winters. MS has warm, relatively short summers and cold winters. ICH is one of the
wettest zones in the interior of BC and has warm dry summers and cool wet winters. ESSF
occurs at high-elevations and has cool and short summers and long, cold, and snowy winters.
The climate of the SBS is characterized as relatively warm, moist, and short summers and
severe, snowy winters. Central Yukon has a continental climate characterized by very cold
and long winters, and warm short summers with relatively low precipitation (Figure 2.2).

Site-selection criteria included 1) more than 60-year old stands to provide sufficient
ring-width records, 2) stands on zonal sites to represent intermediate topographic and edaphic
conditions within an area, 3) the minimal visible evidence of stand-level disturbance events
to minimize non-climatic signals in tree ring-widths, and 4) proximity to a local weather
station to minimize potential errors in site-specific climate records estimated by the
ClimateBC model (described in 2.2.3). Stands were naturally regenerated, mid- to
late-successional stage forests. Sampling along an elevational transect was conducted
wherever possible to provide climate gradients within narrow geographic locations (E1-E4
and N2-N3). Three species were sampled from one stand when possible (i.e., CI, W), but
they were sampled from two or three nearby stands when only single species dominated
sample stands (i.e., S, Nn). Low- to mid- elevation stands (< 1400 m) were characterized as
dense, closed canopy and deciduous-conifer mixed stands. High-elevation stands (1500 1800 m) were typically less dense, open canopy and conifer-dominant stands.

13

2.2.2. Chronology development
One or two tree increment cores were extracted at breast height (1.3 m) from a minimum of
20 trees per species per site in 2005-2006. The coring height of 1.3 m was chosen because it
provided sufficient growth ring records of longer than 50 years. Healthy, canopy-dominant
trees with little observable damage were selected for coring to minimize non-climatic
variation in ring-widths. Standard dendrochronology techniques were applied to develop siteand species-specific tree-ring chronologies (Fritts 1976). Sampled cores were mounted and
sanded with increasing grain size (to 600 per inch) to observe annual rings clearly (Stokes
and Smiley 1968). Crossdating is a procedure to identify the exact year of ring formation by
synchronously matching ring-width patterns among all sampled cores from a given site.
Crossdating helps to detect any missing or false rings, because trees may fail to produce an
annual ring or may produce two rings in a year. Narrow rings were used as pointer-years
during visual crossdating (Yamaguchi 1991). Annual ring-widths were measured to 0.01 mm
using the computer program WinDENDRO™ (Regent Instruments Inc. 2005). The Velmex
ring-measurement system (Velmex Inc. 1992) interfaced with MeasureJ2X (VoorTech
Consulting 2004) with precision 0.001 mm was used to measure small rings. A computer
program COFECHA (Grissino-Mayer 2001) was used to statistically detect potential errors
and to validate visual crossdating. Cores that did not crossdate (with a critical threshold
intercorrelation value of 0.36 based on 40-year segment with 20-year lag) were re-examined
under a microscope and remeasured using MeasureJ2X. Those cores that did not crossdate
after the re-examination were removed from the final chronology development.

The raw ring-width measurements for each crossdated series were standardized by converting
to dimensionless ring-width indices to minimize non-climatic factors and autocorrelations
14

(Fritts 1976). Non-climatic factors influencing ring-width variation may include tree age,
competition, disturbance and random variation (Cook 1985). Younger trees, for instance,
generally have higher photosynthetic rates than older trees. This age-dependent biological
trend often results in the formation of wider annual rings during the early years of growth and
narrower annual rings during the later years regardless of climate condition. Changes in
interspecific and intraspecific competitions and below-ground characteristics also influence
low-frequency variation in ring-widths. Ring-width series may retain autocorrelation because
1) physiological processes within a tree often cause a lag in growth responses to climate and 2)
climate conditions tend to persist from one year to the next.

Each core was standardized by fitting a cubic smoothing spline with a 50% frequency response
cutoff of 20 or 40 years using a program ARSTAN (Cook 1985). The ring-width indices of
individually standardized cores were then averaged among all crossdated cores in a population
to develop a ring-width chronology for each species at each site. The Arstan chronology,
autocorrelation removed and pooled autoregression (common persistence) built back in, was
developed to maximize climate signal in the ring-width variation (Cook and Holmes 1986).
The preliminary results and visual assessment of the standardized chronologies showed that
the 40-year smoothing spline reduced unwanted low-frequency variation while maintaining
strong common signals for most populations. A more stochastic detrending method with the
20-year smoothing spline was used for populations from dense, closed-canopy and
low-elevation stands to minimize low-frequency variation that may have resulted from stand
dynamics (Appendix B). Different spline lengths were applied because standardization should
maximize the climate signals of each chronology from forests with different stand histories and
characteristics. This practice is rare in dendroecology, but has been done to study populations
15

from closed versus open canopy stands (Szeicz 1997).

Mean sensitivity and standard deviation were calculated to evaluate the quality of standardized
chronologies. Mean sensitivity is a measure of response to year-to-year climate variation and
standard deviation indicates low- to medium-frequency variation in the ring-width
chronologies, and higher values indicate that the ring-width chronologies contain common
climate signals (Fritts 1976, Villalba et al. 1994). Mean sensitivity is calculated as the absolute
difference between adjacent ring-widths divided by the mean of the two ring-widths.

2.2.3. Climate data
Site-specific climate data were estimated using a climate model, ClimateBC version 3.2 (Wang
et al. 2006). ClimateBC requires latitude, longitude, and elevation to generate site-specific
monthly, seasonal and annual climate variables in western Canada based on PRISM
(parameter-elevation regressions on independent slopes model). PRISM is a regression-based
model that incorporates geographic influences on climate, including elevation, aspect, coastal
effects and orographic influences (Daly et al. 2002, Hamann and Wang 2005). For example,
elevation is a strong predictor of temperature because temperature decreases almost linearly
with altitude. Based on the topographical features, PRISM adjusts reference weather station
data to estimate mean monthly temperature and precipitation for each grid at a resolution of
approximately 4 km. Information from each reference weather station is weighted based on a
distance from a target grid cell, elevation and other topographic factors to calculate the values
of the target cell. On top of the medium resolution climate data of 4 km grid, ClimateBC uses a
high-resolution digital elevation model to build a locally and temporary scale-free climate
model for finer scale climate estimates (Wang et al. 2006). The verification was conducted by
16

comparing predicted and observed climate data from 191 weather stations that had sufficient
climate records included in the 1951-1980 or 1961-1990 normals (Wang et al. 2006).

2.2.4. Growth-climate relationships
Pearson simple correlation coefficients were calculated to identify climate variable(s)
associated with variations in the Arstan chronologies (P < 0.05). We selected monthly climate
variables for 17 months extending from May of the previous growth year to September of the
current growth year to include two complete growing seasons (Larsen and McDonald 1995,
Brooks et al. 1998). Climate variables used in the correlation analyses included mean,
minimum and maximum monthly temperatures, monthly total precipitation, monthly
heat-moisture indices and annual derived variables, including degree-days, frost-free days and
frost-free period (the number of consecutive frost-free days) (Wang et al. 2006). Heat-moisture
index indicates moisture availability in a given month and is calculated as
Heat-moisture index = temperature / (precipitation / 1000)
Because this index considers both precipitation and evapotranspiration (related to temperature),
it tends to better reflect moisture availability than precipitation alone (Wang et al. 2006).
Degree-days are the amount of heat energy available for plant growth, calculated as the sum of
the difference between mean daily temperature and 5 °C (growing degree-days) or 18 °C
(heating degree-days) for a year. A 50-year period from 1953 to 2002 was used because this
length could minimize errors in climate data estimates related to the small numbers of
monitoring stations in existence prior to 1950.

2.2.5. Gradient analyses
When ring-width chronologies showed strong significant correlations with the climate
17

variables of two or more consecutive months, these monthly climate variables were averaged
to form a seasonal climate variable. Simple linear regression analyses were used to evaluate the
strength and importance of predictor climate variables on variation in ring-width indices. The
slope of regression (bl) represented the direction and strength of the relationship and was
defined as the growth sensitivity of a population to the climate variable. We considered
multiple regression analyses to improve the prediction of annual ring-width indices using
multiple climate variables. However, multiple regression results were not reported because the
predictor climate variables were seldom consistent across the large number of sampled
populations and they would limit comparative investigation across the sites and among species.

The regression coefficients (bl) of populations to the selected predictor climate variables were
examined along climate gradients for each species using simple- and multiple regression
analyses (P < 0.05). Climate gradients used to predict the regression coefficients included
mean annual and seasonal temperatures and precipitation during the 1961-1990 period.
Normality was tested using the Kolmogorov-Smirnov normality test. The analyses were
conducted using the statistical software SPSS (SPSS Inc. 1999).

2.3. Results
2.3.1. Chronology statistics
Forty Arstan ring-width chronologies were developed (Appendix B). Average mean sensitivity
was 0.16 ± 0.03, 0.14 ± 0.04 and 0.12 ± 0.02 for lodgepole pine, interior spruce and subalpine
fir, respectively. Previous studies suggest that a mean sensitivity range of 0,09-0.16 is
sufficient to make growth-climate comparisons for the three species (Villalba et al. 1994, Ettl
and Peterson 1995). Standard deviations were 0.77 ± 0.25, 0.58 ± 0.22 and 0.57 ± 0.18 for
18

lodgepole pine, interior spruce and subalpine fir, respectively.

2.3.2. Growth-climate relationships
We focused on correlations established between ring-width chronologies and mean monthly
temperatures to examine growth-climate relationships for two reasons. First, ring-width
chronologies generally had stronger and more consistent correlations with mean monthly
temperatures than with monthly precipitation (Appendix C). Second, ring-width chronologies
showed significant correlations with several climate variables that were highly associated with
each other. For example, populations that showed positive correlations with mean June-July
temperatures generally showed positive correlations with June-July maximum and minimum
temperatures and growing and heating degree-days. The objective of this study was the
comparative investigation among populations rather than among the related climate variables
and thus, we mainly presented the relationships between growth and monthly temperatures,
except where there were important precipitation and heat-moisture correlations that might be
related to temperature influences.

Our data showed that significant correlations between monthly temperatures and ring-width
chronologies varied among the three species (Table 2.2). Eleven out of the 14 interior spruce
populations had significant correlations with June and/or July temperatures of the current
growth year (Table 2.2a). Interior spruce populations at low-elevation and warmer sites in BC
(S, CI, C2, El, E2) and northern most site in Yukon (Nn) showed either negative or no
correlations with June temperatures of the current growth year. However, high-elevation and
western populations in BC (E3, E4, W, Ww) and two Yukon populations (Nl, N2) were
positively correlated with June and July temperatures.
19

Monthly temperature variables that were significantly correlated with lodgepole pine
ring-width chronologies varied among sites, most notably between BC and Yukon. BC
populations except for the southern most site (S) were predominantly positively correlated
with temperatures from October to March prior to growth, with an exception of no significant
correlation in December (Table 2.2b). Seven out of the 10 BC populations also showed weak
(Appendix C) positive correlations with July and August temperatures of the current growth
year. Yukon populations, however, were primarily negatively correlated with June
temperatures of the current growth year and summer temperatures in the previous growth year.
In addition, Yukon populations also showed negative correlations with summer heat-moisture
conditions, including July and August of the previous growth year and June of the current
growth year (Table 2.3).

Monthly temperature variables that were significantly correlated with subalpine fir varied
considerably among the sample sites (Table 2.2c). Within BC, high-elevation and western
populations (E4, W, Ww) were strongly positively correlated with temperatures from October
to March prior to growth (Table 2.2c, Appendix C). During summer months, western
populations (W, Ww) were positively correlated with July temperatures of the current growth
year, while low- to mid-elevation populations in BC (El-3, CI, S) were negatively correlated
with May and/or June temperatures of the current growth year and July and/or August
temperatures of the previous growth year. Yukon populations were negatively correlated with
temperatures of the previous or current growing seasons.

20

2.3.3. Gradient analyses
Based on the correlations (Table 2.2), we selected growing season and October-March
temperature variables to analyze how sensitivities of populations to these common predictor
variables changed along climate gradients (Table 2.4). Growing season temperature was
defined as any month(s) between May and August with which each population showed
significant correlations. The regression coefficients (bl) of interior spruce and lodgepole pine
to the growing season temperatures changed linearly along gradients in mean summer
(June-August) temperature and/or precipitation (Figure 2.3). The regression coefficients of
lodgepole pine and subalpine fir to October-March temperatures changed along the mean
annual temperature gradient in BC (Figure 2.4). We considered the interactive effects of site
temperature and precipitation to predict regression coefficients, but no multiple regressions
were significant. Thus, only results from simple linear regression were presented.

2.3.3.1. Growing season temperatures
The regression coefficients of interior spruce ring-width indices regressed on June-July
temperatures had an inverse relationship with mean summer temperatures (P < 0.0004, Figure
2.3a), but had no significant relationship with mean summer precipitation (P = 0.51, Figure
2.3b). The threshold temperature (x-intercept at y = 0) where the regression relationships
changed from positive to negative was 12.4 °C.

Important growing season temperature variables that predicted the ring-width indices of
lodgepole pine varied between BC and Yukon (Table 2.4). In BC, the regression coefficients of
lodgepole pine ring-width indices against July-August temperatures had negative relationship
with mean summer temperatures (P = 0.02, Figure 2.3c). No significant correlation was found
21

between the regression coefficients and mean summer temperatures across the entire study
region (P = 0.09, Figure 2.3c). However, a non-linear positive relationship was observed
between mean summer precipitation and the regression coefficients of the ring-width indices
against growing season temperatures (July-August for BC and June for Yukon) across the
entire study region (P < 0.0004, Figure 2.3d). Further examination of the non-linear
relationship revealed that the regression coefficients of pine increased linearly along summer
precipitation up to 250 mm (P < 0.0004), after which no further increases were observed. The
threshold precipitation (x-intercept at y = 0) where significant relationships changed from
positive to negative was 167 mm.

Although eight out of the 12 subalpine fir ring-width chronologies were significantly
correlated with growing season temperatures (either positively to July or negatively to
May-June, Table 2.4), the regression coefficients of fir against the predictor growing season
temperatures did not change along gradients of mean summer temperatures (P = 0.07) or mean
summer precipitation (P = 0.80). No linear tend was found among BC populations for either
summer temperature or precipitation gradients (P > 0.16).

2.3.3.2. October-March temperatures
Based on the correlations found with lodgepole pine and subalpine fir (Table 2.2), mean
monthly temperatures from October to March prior to growing season were averaged. In BC,
nine out of the ten lodgepole pine and five out of the nine subalpine fir ring-width chronologies
had positive correlations with average October-March temperatures prior to growth (Table 2.4).
No such relationship was observed for any Yukon populations or spruce. Within BC, the
regression coefficients of ring-width indices against October-March temperatures had a
22

negative relationship with mean annual temperatures for lodgepole pine (P < 0.0004) and
subalpine fir (P = 0.009, Figure 2.4). The regression coefficients of lodgepole pine and
subalpine fir against October-March temperatures were not correlated with mean annual
precipitation for lodgepole pine (P = 0.71) or fir (P = 0.80) within BC. Mean annual
temperatures were significantly positively correlated with growing degree-days (> 5°C),
frost-free days and frost-free period within BC (P < 0.0004 for each variable).

2.4. Discussion
Our data suggest that 1) the response and sensitivity of trees to seasonal climate variables vary
among species and sites, 2) the growth-climate relationships of interior spruce may differ
distinctively from those of lodgepole pine and subalpine fir examined across the wide
geographic and climate ranges sampled in British Columbia (BC) and Yukon, 3) winter
temperatures prior to active growth may be important in assessing the potential impacts of
climate change on tree growth and forest communities and 4) the shade-tolerance and
successional positions of tree species may not contribute to prediction of the radial
growth-climate relationships for mature trees.

2.4.1. Spatial patterns
The three study species showed different spatial variations in growth-climate relationships
across the sample sites in BC and Yukon. Our data suggest that interior spruce may exhibit a
common ring-width response to summer temperatures over several thousand kilometers in
western Canada. Previous studies also found spatially consistent growth responses (positively
or negatively) to summer temperatures among the populations of white spruce (Peterson and
Peterson 1994), Engelmann spruce (St. George and Luckman 2001) and Norway spruce (Picea
23

abies [L.] Karst.) (Makinen et al. 2000, 2002). For example, St. George and Luckman (2001)
studied 17 Engelmann spruce ring-width chronologies across a 500-km-long sample region in
the Canadian Rocky Mountains, and they found positive correlations with summer
temperatures (most obviously in June-July) of the current growth year for most populations. In
Scandinavia, Makinen et al. (2000) reported the similar results for 40 Norway spruce
populations sampled from five geographic regions, characterized as having different
environmental conditions, extending from central Finland to the Arctic timberline.

Our data showed that the growth-climate relationships of lodgepole pine differed distinctively
between BC and Yukon. Wheeler and Guries (1982) suggest the genetically distinct subgroups
of interior lodgepole pine between BC and Yukon, based on the number and frequency of rare
alleles. They suggest that the genetic differentiation of lodgepole pine possibly occurred during
the most recent glacial event. Lodgepole pine that is currently distributed in western Canada
probably existed in two separate ice-free refugia, including western central Yukon and the
southwestern United States. The northern populations of lodgepole pine might have migrated
southward as the glacier retreated and met by the lodgepole pine populations moving north,
probably around a few kilometers south of the BC-Yukon boarder. Xie and Ying (1995) also
suggest distinct lodgepole pine subgroups based on provenance studies. They reported that
lodgepole pine populations from southern and central BC showed similar variations in 20-year
height growth and volume, but they showed significantly different variations in these growth
traits from populations north of 57 °N. Therefore, genetic differentiation within species may
explain the distinct growth-climate relationships found between BC and Yukon populations in
this study.

24

Our data and previous studies suggest that the growth-climate relationships of subalpine fir
may be spatially variable due to local site conditions (Villalba et al. 1994, Peterson et al. 2002)
and elevation (Splechtna et al. 2000). For example, Peterson et al. (2002) reported that the
growth responses of subalpine fir varied between wet and cool versus dry and warm sites
within a 370-km sample region in western Oregon and Washington. Our data agreed with
previous studies that subalpine fir populations at low-elevation were negatively correlated with
May and/or June temperatures of the current growth year while populations at high-elevations
were positively correlated with July temperatures in the southern and central interior of BC
(Splechtna et al. 2000) and on the northwest coast of the United States (Ettl and Peterson 1995,
Peterson et al. 2002). Several studies also reported that subalpine fir had strong associations
with winter and spring temperatures and snow depth at high-elevations but not at
low-elevations (Ettl and Peterson 1995, Peterson et al. 2002, Larocque and Smith 2005).

2.4.2. Gradient analyses
2.4.2.1. Growing season temperatures
Our data suggest that local summer temperatures may be important in predicting the growth
responses and sensitivities of interior spruce to June-July temperatures (Figure 2.3a). Warm
June-July temperatures enhance tree radial growth at cooler sites for several possible reasons.
These include higher day- and night-time temperatures that enhance photosynthesis and
carbohydrate allocation to the stem (Korner 1998), higher photosynthetically active radiation
available due to more sunny days (Goldblum and Rigg 2005), the extension of the growing
season (Danby and Hik 2007) and/or the interaction of these factors. Savva et al. (2006), for
example, found that the positive correlation coefficients between the radial growth of Norway
spruce and June-July temperatures increased linearly with elevation. They suggest that warm
25

summer temperatures may favour the growth of populations at high-elevation sites, where
growing season is generally short. St. George and Luckman (2001) and Wilson and Luckman
(2003) also reported similar results for high-elevation Engelmann spruce in BC. By contrast,
warm temperatures during the growing season may induce moisture stress and cause reduced
growth at warm sites (Barber et al. 2000, D'Arrigo et al. 2004, 2005, Wilmking et al. 2004).
Warm temperatures often increase evapotranspiration from soils and plant tissues (Brooks et al.
1991). Trees may increase transpiration rates and close a large proportion of stomata on leaves
to minimize water losses in response to warm temperatures and moisture stress, which may
result in reduced net photosynthesis and growth (Kozlowski 2002). For example, Barber et al.
(2000) reported that the ring-width chronologies of low-elevation white spruce were
negatively correlated with summer temperatures in the semi-arid interior of Alaska, where
evapotraspiration potentially equals annual precipitation. They found that the negative
relationship was associated with reduced CO2 uptake and higher water vapour loss during
photosynthesis.

In contrast with spruce, local precipitation may be more important than temperatures in
predicting the growth sensitivities of lodgepole pine to growing season temperatures (Figure
2.3d). All Yukon sites received less than 150 mm of summer precipitation and thus, warm
temperatures might have induced moisture stress on tree radial growth as suggested in spruce
(Barber et al. 2000, D'Arrigo et al. 2004, 2005, Wilmking et al. 2004). On the other hand,
temperature-induced moisture stress was unlikely for lodgepole pine in BC because they
occurred at sites moister than 155 mm and had positive correlations with growing season
temperatures. Among the three study species, lodgepole pine tends to grow at warmer and drier
sites, whereas interior spruce often dominates cooler and moister environments (Burns and
26

Honkala 1990). Therefore, moisture may be more of a limiting resource for lodgepole pine,
and temperature may be more of a limiting resource for interior spruce during the growing
season.

2.4.2.2. October-March temperatures
Temperatures prior to active growth may have stronger impact than growing season
temperatures on the radial growth of lodgepole pine and subalpine fir in BC. Several studies
suggest that the extension of the growing season, associated with early snowmelt and late
snowfall, may explain the positive effects of winter temperatures on tree radial growth at
colder sites (Graumlich and Brubaker 1986, Splechtna et al. 2000, Peterson and Peterson 2001,
Kirdyanov et al. 2003, Pederson et al. 2004, Pfeifer et al. 2005). Peterson and Peterson (2001)
and Peterson et al. (2002) suggest that the timing of snowmelt in the spring may determine the
date of growth initiation of subalpine fir at high-elevations. Several studies showed that
snowpack in early spring (April-May) was negatively correlated with the radial growth of
subalpine fir (Peterson and Peterson 1994, Peterson et al. 2002, Larocque and Smith 2005),
lodgepole pine (Case and Peterson 2007) and mountain hemlock (Graumlich and Brubaker
1986, Peterson and Peterson 2001) at high-elevations in western North America and several
Larix species in subarctic Russia (Kirdyanov et al. 2003).

Preconditioning may explain the positive influences of warm early winter temperatures on tree
growth in the following growing season (Lebourgeois 2000). Although stem growth processes
cease in late summer, trees may continue photosynthetic activities under favourable growing
conditions that are extended into late fall. Carbohydrates produced at this time can be stored in
plant tissues (i.e., roots, twigs, old leaves) in the fall and used for stem growth in the following
27

spring (Kramer and Kozlowski 1979, Pfeifer et al. 2005). Previous studies, for example,
showed that previous fall temperatures were positively correlated with the radial growth of
Jack pine {Pinus banksiana [Lamb.]) in western Quebec (Hofgaard et al. 1999) and stone pine
(Pinus cembra L.) in the southern European Alps (Oberhuber 2004, Pfeifer et al. 2005, Carrer
et al. 2007).

Our data showed that the importance of warm October-March temperatures on radial growth
increased as mean annual temperatures decreased. Mean annual temperatures were positively
correlated with growing degree-days (> 5 °C), frost-free days and frost-free period within BC,
suggesting that growing season conditions were limited at colder sites in BC. In addition, no
pine populations had significant correlations with temperatures in December (Table 2.2b),
suggesting that temperatures in the early and late winter months might be more important than
those in the mid-winter for lodgepole pine.

The potential negative impacts of high snowfall on tree growth (Peterson and Peterson 2001,
Larocque and Smith 2005) remained unclear in this study because our data did not show
significant relationships between winter precipitation and the ring-width chronologies of
lodgepole pine or subalpine fir (Appendix C). However, Knowles et al. (2006) showed that
reduced snowfall in the western United States in the past 55 years was unrelated to the changes
in the total precipitation. Holding precipitation constant, warm winter temperatures decrease
the percentage of precipitation that falls as snow and increase dense, wet snow that is close to
melting point (MWLAP 2002, Wang et al. 2006). Therefore, variation in temperature rather
than precipitation may have more impacts on the total amounts of snowfall during winter and
consequently the timing of snowmelt in spring (Knowles et al. 2006). Actual snow data may
28

help in examining the potential influences of snow on tree radial growth.

2.4.3. Growth responses of ecologically distinct species
This study suggests that shade-tolerance or successional positions may not contribute to
predict the growth responses of coexisting species to climate variables across their ranges in
BC and Yukon, at least for mature trees growing under natural conditions. Interior spruce and
subalpine fir are expected to exhibit similar growth responses to climate because they often
co-dominate cool and wet habitats and show more shade-tolerance and late-successional
positions than lodgepole pine (Burns and Honkala 1990, Green 2005, 2007). In this study, the
growth responses of interior spruce differed distinctively from those of lodgepole pine and
subalpine fir. By contrast, Green (2005, 2007) found distinct clinal patterns in phenological
responses between lodgepole pine and subalpine fir seedlings along an elevational transect in
central BC. These contradictory observations between our data and previous studies may be
related to tree age. Tree physiology generally changes with age, which may cause shifts in
sensitivities to environmental stresses (Carrer and Urbinati 2004). For example, He et al.
(2005) reported that the drought sensitivities of four deciduous tree species were associated
with shade-tolerance and successional positions in saplings, but not in mature trees. Van Der
Kamp (1990) also reported that the susceptibility of conifer buds to the extreme cold weather
differed between young (10-15 years) and mature trees in BC. Differences in growth responses
among species with different levels of shade-tolerance and successional positions may be more
directly related to changes in competition levels (i.e., light) and the frequency of disturbance
(i.e., fire), respectively, than to climate (Johnstone and Chapin 2003, Simard et al. 2004).

Biogeography and genecology of each species may provide possible explanations of the
29

species- and site-specific adaptive responses to climate. For example, lodgepole pine in Yukon
and BC exhibited unique responses to climate possibly due to distinct genetic structures and
migration routes found between the two subgroups (Wheeler and Guries 1982, Xie and Ying
1995). Similarly, geologic studies suggest that subalpine fir existed in the coastal regions of the
southwestern United States (Ettl and Peterson 2001) while white spruce existed in the
unglaciated regions in Yukon and Alaska (Cwynar and Spear 1991) during the most recent
glacial period. These two species might have been subjected to different selection pressures
during their migrations into the current ranges in BC and Yukon from different routes. White
spruce was probably more suited for cold continental climate than subalpine fir (Cwynar and
Spear 1991). Therefore, evolutionary history might explain the expression of unique
growth-climate relationships among species.

2.4.4. Implications
Different growth responses to seasonal climate among coexisting species may alter the species
interaction, dominance and composition of forest communities under future climate change
(Stohlgren and Bachand 1997, Zolbrod and Peterson 1999, Millar et al. 2004). For example,
changes in growing season temperatures may significantly alter competitive relationships
between lodgepole pine and interior spruce at warmer sites in BC, because lodgepole pine was
positively and interior spruce was negatively correlated with growing season temperatures at
sites warmer than 12.4 °C. This study also suggests that changes in winter temperatures can
have significant impacts on tree radial growth and forest ecosystems. Instrumental climate
records showed that the greatest magnitude of warming in the past 150 years occurred in winter
and early spring, rather than summer, at the global scale (Jones et al. 1999, Barber et al. 2000).
General circulation models also predict that winter minimum temperatures will increase and
30

more winter precipitation will occur as rain in northwestern North America in the next several
decades (Christensen 2007). Warm winter temperatures may favour the growth of lodgepole
pine and subalpine fir at cold sites in the interior of BC, potentially resulting in higher survival
rates, mechanical stability and reproductive capacity of these species (Givnish 1995, Despland
and Houle 1997, Loehle 2000). Consequently, lodgepole pine may increase in abundance and
shift its ranges into high-elevation spruce-fir forests (Millar et al. 2004). Winter condition,
therefore, should be considered when evaluating the potential impacts of tree growth and forest
ecosystems.

31

2.5. Literature cited
Barber, V.A., Juday, G.P., and Finney, B.P. 2000. Reduced growth of Alaskan white spruce in
the twentieth century from temperature-induced drought stress. Nature. 405: 668-673.
Bazzaz, F.A. 1987. Experimental studies on the evolution of niche in successional plant
populations. In Colonization, succession and stability. Edited by A.J. Gray, M.J.
Crawley, P.J. Edwards. Blackwell Scientific, Oxford, pp 245-271.
Bertrand, A., and Castonguay, Y. 2003. Plant adaptations to overwintering stresses and
implications of climate change. Canadian Journal of Botany. 81: 1145-1152.
Brooks, K.N., Ffolliott, P.F., Gregersen, H.M., and Thames, J.L. 1991. Hydrology and the
management of watersheds. 1st Edition. Iowa State University Press. Iowa, pp 37-63.
Brooks, J.R., Flanagan, B., and Ehleringer, J.R. 1998. Responses of boreal conifers to climate
fluctuations: indications from tree-ring widths and carbon isotope analyses. Canadian
Journal for Forest Research. 28: 524-533.
Burns, R.M., and Honkala, B.H. 1990. Silvics of North America: 1. Conifers; 2. Hardwoods.
Agriculture Handbook 654. U.S. Department of Agriculture, Forest Service,
Washington, DC. vol.2, 877 pp.
Carrer, M., Nola, P., Louis, E., Motta, R., and Urbinati, C. 2007. Regional variability of
climate-growth relationships in Pinus cembra high elevation forests in the Alps. Journal
of Ecology. 95: 1072-1083.
Carrer, M., and Urbinati, C. 2004. Age-dependent tree-ring growth responses to climate in
Larix deciduas and Pinus cembra. Ecology. 85: 730-740.
Case, M.J., and Peterson, D.L. 2007. Growth-climate relationships of lodgepole pine in the
North Cascade National Park, Washington. Northwest Science. 81: 62-75.
Christensen, J.H., Hewitson, B., Busuioc, A., Chen, A., Gao, X., Held, I., Jones, R., Kolli, R.K.,
Kwon, W.-T., Laprise, R., Magafia Rueda, V., Mearns, L., Menendez, C.G., Raisanen,
J., Rinke, A., Sarr A., and Whetton, P. 2007. Regional Climate Projections. In: Climate
Change 2007: The Physical Science Basis. Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and
H.L. Miller (eds.). Cambridge University Press, Cambridge, United Kingdom and New
York, NY, USA.
Cook, E.R. 1985. A time series analysis approach to tree ring standardization. Ph.D.
dissertation, University of Arizona, Tucson.
Cook, E.R., and Holmes, R.L. 1986. Users Manual for Program ARSTAN. Laboratory of
Tree-ring Research, University of Arizona, Tucson, USA.
32

Cook, E.R., and Cole, J. 1991. On predicting the response of forests in eastern North America
to future climatic change. Climatic Change. 19: 271-282.
Cwynar, L.C., and Spear, R.W. 1991. Reversion of forest to tundra in the central Yukon.
Ecology. 71:202-212.
D'Arrigo, R.D., Kaufmann, R.K., Davi, N., Jacoby, G.C., Laskowski, C, Myneni., R.B., and
Cherubini, R 2004. Thresholds for warming-induced growth decline at elevational
tree line in the Yukon Territory, Canada. Global Biogeochemical Cycles. 18: GB3021,
doi: 10.1029/2004GB002249.
D'Arrigo, R.D., Mashig, E., Frank, D., Wilson, R., and Jacoby, G. 2005. Temperature
variability over the past millennium inferred from Northwestern Alaska tree rings.
Climate Dynamics. 24: 227-236.
Daly, C, Gibson, W. R, Taylor, G. H., Johnson, G.L., and Pasteris, R 2002. A knowledge-based
approach to the statistical mapping of climate. Climate Research 22: 99-113.
Danby, R.K., and Hik, D.S. 2007. Responses of white spruce (Picea glauca) to experimental
warming at a subarctic alpine treeline. Global Change Biology. 13: 437-451.
Despland, E., and Houle, G. 1997. Climate influences on growth and reproduction ofPinus
banksiana (Pinaceae) at the limit of the species distribution in eastern North America.
American Journal of Botany. 84: 928-937.
Ettl, G.J., and Peterson, D.L. 1995. Growth-response of sub-alpine fir (Abies lasiocarpa) to
climate in the Olympic Mountains, Washington, USA. Global Change Biology. 1:
213-230.
Ettl, G.J., and Peterson, D.L. 2001. Genetic variation of subalpine fir (Abies lasiocarpa (Hook.)
Nutt.) in the Olympic Mountains, WA, USA. Silvae Genetica. 50: 145-153.
Fritts, H.C. 1976. Tree rings and climate. The Blackburn Press. Caldwell, NJ. 567 pp.
Givnish, T.J. 1995. Plant stems: biomechanical adaptation for energy capture and influence on
species distributions. In Plant stems: physiology and functional morphology. Edited by.
G.L. Barbara. Academic Press. Dan Diego, California, pp 3-49.
Goldblum, D., and Rigg, L.S. 2005. Tree growth response to climate change at the
deciduous-boreat forest ecotone, Ontario, Canada. Canadian Journal of Forest
Research. 35:2709-2718.
Graumlich, L.J. 1993. Response of tree growth to climatic variation in the mixed conifer and
deciduous forests of the upper Great Lakes region. Canadian Journal of Forest
Research. 23: 133-143.

33

Graumlich, L.J., and Brubaker, L.B. 1986. Reconstruction of Annual Temperature
(1590-1979) for Longmire, Washington, Derived from Tree Rings. Quaternary
Research. 25: 223-234.
Green, D.S. 2005. Adaptive strategies in seedlings of three co-occurring, ecologically distinct
northern coniferous tree species across an elevational gradient. Canadian Journal of
Forest Research. 35: 910-917.
Green, D.S. 2007. Controls of growth phenology vary in seedlings of three, co-occurring
ecologically distinct northern conifers. Tree Physiology. 27: 1197-1205.
Grissino-Mayer, H.D. 2001. Evaluating crossdating accuracy: a manual and tutorial for the
computer program COFECHA. Tree-Ring Research. 57: 205-221.
Hamann, A., and Wang, T.L. 2005. Models of climatic normals for genecology and climate
change studies in British Columbia. Agricultural and Forest Meteorology. 128:
211-221.
Hamrick, J.L. 2004. Response of forest trees to global environmental changes. Forest Ecology
and Management. 197: 323-335.
Hanninen, H., Beuker, E., Johnsen, 0., leinonen, I., Murray, M., Sheppard, L. and Skroppa, T.
2001. Impacts of climate change on cold hardiness of conifers. In Conifer cold
hardiness. Edited by F.J. Bigras and S.J. Colombo. Kluwer Academic Publishers,
Dordrecht, Netherlands, pp 23-53.
He, J-S., Zhange, Q-B., and Bazzaz, F.A. 2005. Differential drought responses between
saplings and adult trees in four co-occurring species of New England. Trees. 19:
442-450.
Hofgaard, A., Tardif, J. and Bergeron, Y. 1999. Dendroclimatic response of Picea mariana and
Pinus banksiana along a latitudinal gradient in the eastern Canadian boreal forest.
Canadian Journal of Forest Research. 29: 1333-1346.
Johnstone, J.F., and Chapine, F.S. 2003. Non-equilibrium succession dynamics indicate
continued northern migration of lodgepole pine. Global Change Biology. 9:
1401-1409.
Jones, P.D., New, M., Parker, D.E., Martin, S., and Rigor, I.G. 1999. Surface air temperature
and its changes over the past 150 years. Reviews of Geophysics. 37: 173-199.
Kirdyanov, A., Hughes, M., Vaganov, E., Schweingruber, F., and Silkin, P. 2003. The
importance of early summer temperature and date of snow melt for tree growth in the
Siberian Subarctic. Trees. 17: 61-69.
Knowles, N., Dettinger, M.D., and Cayan, D.R. 2006. Trends in snowfall versus rainfall in the
Western United States. Journal of Climate. 19: 4545-4559.
34

Korner, C. 1998. Are-assessment of high elevation treeline positions and their explanation.
Oecologia. 115:445-459.
Kozlowski, T.T. 2002. Acclimation and adaptive responses of woody plants to environmental
stresses. The Botanical Review. 68: 270-334.
Kramer, K., Leinonen, I., and Loustau, D. 2000. The importance of phenology for the
evaluation of impact of climate change on growth of boreal, temperate and
Mediterranean forests ecosystems: an overview International Journal of
Biometeorology. 44: 67-75.
Kramer, P.J., and Kozlowski, T.T. 1979. Physiology of woody plants. Academic Press. N.Y.
USA. 811pp.
Larocque, S.J., and Smith, D.J. 2005. A dendroclimatological reconstruction of climate since
AD 1700 in the Mt. Waddington area, British Columbia Coast Mountains, Canada.
Dendrochronologia. 22: 93-106.
Larsen, C.P.S., and MacDonald, G.M. 1995. Relations between tree-ring widths, climate, and
annual area burned in the boreal forest of Alberta. Canadian Journal of Forest
Research. 25: 1746-1755.
Lebourgeois, F. 2000. Climatic signals in early wood, latewood and total ring width of
Corsican pine from western France. Annals of Forest Science. 57: 155-164.
Linderholm, H.W. 2001. Climatic influence on scots pine growth on dry and wet soils in the
central Scandinavian mountains, interpreted from tree-ring widths. Silva Fennica. 35:
415-424.
Loehle, C. 2000. Strategy space and the disturbance spectrum: a life-history model for tree
species coexistence. American Naturalist. 156: 14-33.
Makinen, H., Nqjd, P., and Mielikainen, K. 2000. Climatic signal in annual growth variation of
Norway spruce {Picea abies) along a transect from central Finland to the Arctic
timberline. Canadian Journal of Forest Research. 30: 769-777.
Makinen, H., Nqjd, P., Kahle, H.P., Neumann, U., Tveite, B., Mielikainen, K., Rohle, H., and
Spiecker, H. 2002. Radial growth variation of Norway spruce {Picea abies (L.) Karst.)
across latitudinal and altitudinal gradients in central and northern Europe. Forest
Ecology and Management. 171: 243-259.
May, R.M. 1974. On the theory of niche overlap. Theoretical Population Biology. 5: 297-332.
Meidinger, D., and Pojar, J. 1991. Ecosystems of British Columbia. BC. Ministry of Forests.
Victoria, BC. 330 pp.

35

Millar, C.I., Westfall, R.D., Delany, D.L., King, J.C., and Graumlich, L.J. 2004. Response of
subalpine conifers in the Sierra Nevada, California, U.S.A., to 20th-century warming
and decadal climate variability. Arctic, Antarctic, and Alpine Research. 36: 181-200.
MWLAR Ministry of Water, Land and Air Protection, British Columbia. 2002. Indicators of
climate change for British Columbia, 2002. Victoria, BC. [available online]
http:www.gov.BC.ca/wlap. 48 pp.
Oberhuber, W. 2004. Influence of climate on radial growth of Pinus cembra within the alpine
timberline ecotone. Tree Physiology. 24: 291-301.
Pederson, N., Cook, E.R., Jacoby, G C , Peteer, D.M., and Griffin, K.L. 2004. The influence of
winter temperatures on the annual radial growth of six northern range margin tree
species. Dendrochronologia. 22: 7-29.
Peterson, D.W., and Peterson, D. L. 1994. Effects of climate on radial growth of subalpine
conifers in the North Cascade Mountains. Canadian Journal of Forest Research. 24:
1921-1932.
Peterson, D.W., and Peterson, D. L. 2001. Mountain hemlock growth responds to climatic
variability at annual and decadal time scales. Ecology. 82: 3330-3345.
Peterson, D.W., Peterson, D.L., and Ettl, G. J. 2002. Growth responses of subalpine fir to
climatic variability in the pacific Northwest. Canadian Journal of Forest Research. 32:
1503-1517.
Pfeifer, K., Kofler, W., and Oberhuber, W. 2005. Climate related causes of distinct radial
growth reductions in Pinus cembra during the last 200 years. Vegetation History and
Archaeobotany. 14: 211-220.
Regent Instruments Inc. 2005. WinDENDRO™. An image analysis system for tree-rings
analysis. Regent Instruments Inc. Quebec, Canada.
Sarr, D.A., Hibbs, D.E., and Huston, M.A. 2005. A hierarchical perspective of plant diversity.
The Quarterly Review of Biology. 80: 187-212.
Savva, Y., Oleksyn, J., Reich, P.B., Tjoelker, M.G., Vaganov, E.A., and Modrzynski, J. 2006.
Interannual growth response of Norway spruce to climate along an altitudinal gradient
in the Tatra Mountains, Poland. Trees. 20: 735-746.
Simard, S.W., Sachs, D.L., Vyse, A., and Blevins, L.L. 2004. Paper birch competitive effects
vary with conifer tree species and stand age in interior British Columbia forests:
implications for reforestation policy and practice. Forest Ecology and Management.
198: 55-74.
Spittlehouse, D.L. 2005. Integrating climate change adaptation into forest management.
Forestry Chronicle. 81: 691-695.
36

Splechtna, B.E., Dobry, J., and Klinka, K. 2000. Tree-ring characteristics of subalpine fir
(Abies lasiocarpa (Hook.) Nutt.) in relation to elevation and climatic fluctuations.
Annals of Forest Science. 57: 89-100.
SPSS Inc. 1999. SPSS. Version 15.0. Chicago, 111.
St. George, S., and Luckman, B.H. 2001. Extracting a paleotemperature record from Picea
engelmannii tree-line sites in the central Canadian Rockies. Canadian Journal of
Forest Research. 31: 457-470.
Stohlgren, T.J., and Bachand, R.R. 1997. Lodgepole pine (Pinus contortd) ecotones in Rocky
Mountain National Park, Colorado, USA. Ecology. 78: 632-641.
Stokes, M.A., and Smiley, T.L. 1968. An introduction to tree-ring dating. The University of
Chicago Press. 73 pp.
Szeicz, J.M. 1997. Growth trends and climatic sensitivity of trees in the North Patagonian rain
forest of Chile. Canadian Journal of Forest Research. 27: 1003-1014.
Yamaguchi, D.K. 1991. A simple method for crossdating increment cores from living tree.
Canadian Journal of Forest Research. 21: 414-416.
Van Der Kamp, B.J., and Worrall, J. 1990. An unusual case of winter bud damage in British
Columbia interior conifers. Canadian Journal of Forest Research. 20: 1640-1647.
Velmex Inc. 1992. The Velmex "TA" System for research and non-contact measurement
analysis. Velmex Inc., Bloomfield, N.Y.
Villalba, R., Veblen, T.T., and Ogden, J. 1994. Climatic influences on the growth of subalpine
trees in the Colorado Front Range. Ecology. 75: 1450-1462.
VoorTech Consulting. 2004. MeasureJ2X. VoorTech Consulting, Holderness, N.H.
Walther, G-R. 2003. Plants in a warmer world. Perspectives in Plant Ecology, Evolution and
Systematics. 6: 169-185.
Wang, T., Hanann, A., Spittlehouse, D.L., and Aitken S.N. 2006. Development of scale-free
climate data for western Canada for use in resource management. International Journal
of Climatology. 26: 383-397.
Wheeler, N.C., and Guries, R.P. 1982. Biogeography of lodgepole pine. Canadian Journal of
Botany. 60: 1805-1814.
Wilmking, M., Juday, G.P., Barber, V.A., and Zald, H.J. 2004. Recent climate warming forces
contrasting growth responses of white spruce at treeline in Alaska through
temperature thresholds. Global Change Biology. 10: 1724-1736.

37

Wilson, R.J.S., and Luckman, B.H. 2003. Dendroclimatic reconstruction of maximum summer
temperatures from upper treeline sites in Interior British Coumbia, Canada. Holocen.
13:851-861.
Xie, C-Y, and Ying, C.C. 1995. Genetic architecture and adaptive landscape of interior
lodgepole pine (Pinus contorta spp. latifolia) in Canada. Canadian Journal of Forest
Research. 25:2010-2021.
Zolbrod, A, N., and Peterson, D.L. 1999. Response of high-elevation forests in the Olympic
Mountains to climate change. Canadian Journal of Forest Research. 29: 1966-1978.

38

Table 2.1. Descriptions of the sample sites.
Site

British Columbia
Vernon
Silver Star
King Eddy
King Eddy
Prince George
Cranbrook Hill
Domano Blvd.
McBride
McBride Peak
McBride Peak
McBride Peak
McBride Peak*
Smithers
Onion Mnt. *
Onion Mnt.
Hudson Bay Mnt.
South of Fraser Lake
Top Lake*
Yukon
Whitehorse
Wolf Creek
Grey Mountain
Grey Mountain
Grey Mountain
Mayo, Keno
Mayo
Mayo
Keno Hill

Species

PI

Elevation

Latitude

M

Longitude

m

Aspect

BECin
BC

Site Codes

50° 19'
50° 10'
50° 10'

119° 08'
119° 11'
119° 11'

S
SE
SE

IDF mw

Bl

1000
1165
1200

PVSx/Bl
Pl/Sx

753.5
650.4

53°55'
53° 48'

122°53'
122° 44'

Flat
Flat

SBSdw
SBSmk

CI
C2

Pl/Sx/Bl
Pl/Sx/Bl
Pl/Sx/Bl
Pl/Sx/Bl

1200
1400
1600
1800

53° 19'
53° 19'
53° 19'
53° 20'

120° 09'
120° 08'
120° 07'
120° 07'

SW
SW
SW
SW

ICHmm
ICHmm
ESSFmm
ESSFmm

El
E2
E3
E4*

Pl/Sx/Bl
Sx/Bl
PI

1550
1360
1400

54° 48'
54° 48'
54° 45'

126°52'
126°53'
127°15'

SW
SW

s

ESSF mc
ESSF mc
ESSF wv

Wwl*
Ww2
Ww2

Pl/Sx/Bl

1640

53° 16'

125°10'

SW

ESSF mv

W*

Pl/Sx/Bl
Pl/Sx/Bl
PI
Sx

950
1150
820
845

60° 35'
60° 39'
60° 41'
60° 41'

135°03'
134°53'
134°58'
134°57'

SW
E
E
E

N/A
N/A
N/A
N/A

Nl
N2
N3
N3

Sx
PI
Bl

520
510
1300

63° 37'
63° 29'
63°55'

135°53'
136°16'
135°15'

SW
Flat
S

N/A
N/A
N/A

Nn
Nn
Nn

Sx

MS dm
MS dm

S
S

s

Note: Abbreviations for the tree species are lodgepole pine (PI), interior spruce (Sx) and
subalpine fir (Bl). Abbreviations for biogeoclimatic zones are Interior Douglas-fir (IDF),
Montane Spruce (MS), Interior Cedar-Hemlock (ICH), Engelmann Spruce-Subalpine Fir
(ESSF), and Sub-Boreal-Spruce (SBS); subzones are moist-mild (mm), wet-cool (wk),
wet-cold (wc), dry-warm (dw) and moist-cool (mk) (Meidinger and Pojar 1991). * indicate
altitudinal treeline sites. Site codes were assigned based on relative NEWS direction from
Prince George (Central).

39

Table 2.2. Significant correlation relationships between mean monthly temperatures and
ring-width chronologies from May of the previous growth year to September of the current
growth year for the period 1953-2002. Only significant relationships are shown (P < 0.05).
a) Interior spruce
Current year
Previous year
J F M AM
Site M J J A S 0 N D
s
CI
C2
El
- E2
E3
E4
W
Wwl
+
+
+
+
Ww2
- Nl
N2
+
N3
Nn
- - 0 2 2 0 4
0 5 3 2 1 1 1 0
Number of significant correlations
b) Lodgepole pine
Previous year
Site M J J A S 0 N D

s
CI
C2
El
E2
E3
E4
W
+
Wwl
Ww2 + + +
Nl
.
N2
N3
Nn
1 3 4

+
+
+
+ +
+
+ +
+ +
+
+ +

-

.
-

J

J A S

+
+
+
+
+
+
+

+
+
+
+
+
+
+

11 7 0 0

Current year
J F M AM J
+
+ + +
+ + +
+ +
+ + +
+ +
+
+
+ + +
+
+

J A S
+ +
+ +
+
+ +
+
+

.
+

6 2 5 8 1
7 7 7 0 0
Number of significant correlations

-

+
4 5 5 3

c) Subalpine fir
Previous year
Site M J J A S O N D
S
CI
El
- +
- +
E2
E3
E4
+ + + +
W
+ + +
+ + + +
Wwl + + +
+ +
Ww2
- Nl
N2
- Nn
1 2 5 4 2 5 5 3

Current year
J F M AM J
- - +
+
+
+
+ +

-

J A S
+

+
+ +
+

-

3 1 2 1 4

+
5 5 1 1

Number of significant correlations

40

Table 2.3. Correlation relationships between monthly heat-moisture chronologies and the
ring-width chronologies of lodgepole pine populations in Yukon from May of the previous
growth year to September of the current growth year for the period 1953-2002. Only
significant relationships are shown (P < 0.05).
Previous year
Current year
Site M J J A S 0 N D J F M A M
Nl
N2
- N3
- +
Nn
- +

J J A S
-

Table 2.4. The coefficient of determination (R2) from the simple linear regression analyses
between ring-width chronologies and selected predictor climate variables. Only significant
relationships are shown (P < 0.05).

Sites
S
CI
C2
El
E2
E3
E4
W
Wwl
Ww2
Nl
N2
N3
Nn

Spruce
Growing
season
0.11
0.13
0.17
0.20
0.34
0.48
0.54
0.19
0.16
0.18

Lodgepole pine
Growing
Oct- Summer
season
March
HM
0.18
0.21
0.16
0.18 a
0.25
0.16 a
a
0.34
0.11
a
0.23
0.08
a
0.24
0.11
0.18
0.09 b
0.15
0.14 b
0.19
0.18
0.13 c
0.24
0.12 c
0.22
0.11 c

Subalpine fir
Growing
Octseason
March
0.30 d
n/a
n/a
0.23 d
0.15 d
d
0.10
0.09
0.25
b
0.25
0.16
0.42
0.19 b
0.20
0.11 b
0.14 d
n/a
n/a
-

Note: a growing season = July-August, b growing season = July, c growing season = June,
d
growing season = May-June. Growing season for all spruce populations are June-July,
n/a = no chronologies were established for the sites. Summer heat-moisture (HM) is the
average summer heat-moisture from the previous and current growth year. The significance of
the regressions are P< 0.001 (R2> 0.19), P = 0.001-0.01 (R2 = 0.13-0.19), P> 0.01 (R2<
0.13). Shaded values indicate negative relationships.

41

Nn

Yukon

\

British Columbia

Figure 2.1. Study locations in Yukon and British Columbia, Canada. Site codes were assigned
based on relative NEWS direction from Prince George (Central).

42

1400 -

E4
E3

1200 -

E2
E1

1000 800

^kw2

600

C1(S2

Nn (Keno)
400 -

Nn (Mayo)

N2

-4.0

-2.0

%

200 -

-6.0

0.0

2.0

4.0

6.0

Mean Annual Temperature ( C)

Figure 2.2. The range of climate conditions across the sample sites (ClimateBC version 3.2).
Each point represents the mean annual temperature and mean annual precipitation of each
sample site for the period 1961-1990.

43

3
;«
.1
i£

0.08
0.06
0.04
0.02

J

0.00

-0.025x +0.31
R2-0.70

•A

3
8

0.08
0.06

S -0.02
o
1 -0.04

S
j°
5
o
1

&-0.06

6 0.06

oi -0.08
C

0.08

I

0'04

A

J 0.04

(a) Spruce

0.02
0.00
0.02

A

•

0.04

B! 0.08

(b) Sriru

"f- °'06

IB °' 0 2
g 0.00
u
-0.02
c
• 2 -0.04

A

A

y = 0.089Ln(x)-0.46
R2 0.63

(c) F i n e
9

10

11

12

13

Summer Temperature (°C)

14

15

150

200

250

300

350

40(

Summer Precipitation (mm)

Figure 2.3. The growth sensitivities (regression coefficients: bl) of interior spruce and
lodgepole pine to the current growing season temperatures (June-August) along summer
temperature and precipitation gradients. The following symbols • and A represent BC and
Yukon chronologies, respectively. Value zero indicates that the standardized chronology had
no significant correlations with growing season temperature variables. The solid lines
represent the broad trends across the sample sites (P < 0.05). The dotted lines represent (c) the
regional trend in BC (y = -0.01 lx + 0.18, R2 = 0.77), and (d) the trend across the sites below the
precipitation threshold of 250 mm (y = 0.0012x - 0.20, R2 = 0.80). Site summer temperature
and precipitation were the 1961-1990 normal.

44

^

0.06

&

0.05

(a) Lodgepole pine

y = -0.0081x + 0.0511
R2 = 0.62

9

j§* 0.04
s

0.03

O

1§> 0.01
°- 02

^ ^ \
-

0
_ . 0.06

•

•

A AA

y=-0.0153Ln(x) +0.0161
R2 = 0.87

(b) Subalpine fir

S 0.05

•

_§• 0.04
a

0.03

O

'§ 0.02
& 0.01
*

* \
•

0

•

• r^-^»

A

-0.01

-5

-4

-3

-2

-1

()

1

2

3

4

Mean Annual Temperature (oC)

Figure 2.4. The growth sensitivities (regression coefficients: bl) of lodgepole pine (a) and
subalpine fir (b) to October- March temperatures prior to growth along the mean annual
temperature (°C) gradient. The following symbols • and A represent BC and Yukon
chronologies, respectively. Value zero indicates that the standardized chronology had no
significant correlations with October-March temperatures. Linear and natural logarithmic
regressions were only applied for BC sites. The mean annual temperature of each site was the
1961-1990 normal.

45

Appendix A. Mean annual temperature and mean annual precipitation of each sample site for
the period 1961-1990. SD stands for standard deviation.
Site
Vernon
Prince George
McBride

Smithers
Top Lake
Whitehorse

Mayo
Keno

Site
code
S
CI
C2
El
E2
E3
E4
Wwl
Ww2
W
Nl
N2
N3
Nn
Nn

Annual
Temperature
Mean
SD
4.1
0.85
0.94
3.6
4.1
0.93
2.3
0.89
1.7
0.89
1.1
0.88
0.5
0.88
0.3
0.89
1.0
0.89
0.2
0.88
-1.1
1.26
-1.7
1.27
-1.0
1.27
-3.6
1.41
-3.7
1.40

Annual
Precipitation
Mean
SD
552
84.58
583
72.35
576
68.70
1000 118.87
1103 131.15
1203 143.10
1301 155.04
657
87.49
630
83.76
722 107.35
353
53.60
54.42
359
310
46.87
353
48.71
513
76.16

s

3

•a
C3

?92 2 2 2

O
s 2 2 2

a.

Silver Star
ranbrook H
lomano Blv
ride Peak 1
ride Peak 1
ride Peak 1
ride Peak 1
ion Mountain
dson Bay M
Top Lake
WolfCreefc
Mountain 1 50m
Mountain
Mayo
King Eddy
ranbrook H
Domano Blv
ride Peak 1
ride Peak 1
ride Peak 1
ride Peak 1
ion Mountain
ion Mountain
Top Lake
Wolf Creek
Mountain 1
Grey Mountain 84:
Mayo
King Eddy
ranbrook H
ride Peak 1
ride Peak 1

£

2 2
a

CO

o

a

-3

o

4s. to
O

ON

o

LO

LA

LA

O

o o

LO

to

•z 3

LA
LA

o

o

o

ON

oo
4^ to Oo o O o

QQ
•<

oo
to

£

B
r*

©

o o o
o o3 3
3 3

3

3 3
Wwl

12

o 03 o o
ft)

LA
LA

©

oo
©
©

ON

O
©

CO CO

4^

©
©

to
©
©

a O

I

o-

o

3 3 3 3

3

1 2

M O

LO

o GO

IO

3

•z

to

2 ^

Wwl
Ww2

E o CO

c o

co 03 03 co

O

3 3

3 3

M
to

to

M tfl
4*- LO to

2 Oto o to

ii s.
1/5

o

CO

00

00

00

to

to to to
O O o

o

00

o

LO

to to to

IO
O

tO

to
LA

Lo to LO to to to to
to to
oo o o ^1 LA LO IO to - j ON

U>
00

to to

to
NO

-J

~J

ON

^J

LO

©

to to
4^ ©

to
©

to to

IO

©

ON

LO
00

a. e

o

BLA

IO
4>

to
4> to

00

•O

to to
to to
LO

^i

to
to

00

to
to
4^ © u>

LA

00

IO
©

ON

to
©

K)
©

to to to
© ©

to
ON

>

LO

&n

NO

NO

00

-J

00

NO

00

oo

NO

NO

NO

NO

NO

NO

00

00

NO

oo

NO

NO

oo

00

00

NO

NO

NO

NO

NO

NO

NO

00 00 NO NO 00
ON •o. ©
00
LO ^1 LA NO ©
4*.
to
to © © © ~-J
o to to to
^1
*. to o LA LA 00 NO LO o -o o o -4 © to 00 ON LA -J NO ^1 © OO NO 4^ l-- 00 >-- 4>- ~^tb
to to to to to to to tb to to to to to to to to to to to to to to tb tb tb tb tb tb tb tb tb
o o o O o o o o o o o © © ©© o o ©© o o ©o o © © © © ©© © © © ©
o o o
©
o LA o o o © 4>- © o
o 4^ o
© © © ©
© © o ©
o
o o4*. 4
o*. o
o 4^
Ji -fc. LA o
LA LA o
LA LA LA 4i 4=- 4^
4*. o
44^ 4^ 4^ 4^
4^ LA 4^ 4^
4^ LA

*-

* •

o

S3
O,

cT

*>

o
00

ON

ON

to

ON
00

NO
NO
LA

oo -J
o
£ oo to
4*. O N Lo LA

00

NO
ON

LA

-J

ON

*-* —1 4^ NO

p

O

ON

4^ 4^ LO

NO

LA
to

)° •-^

oo

-o

-J
to

107

o\
LA
ON

©

H^

p p p p p p p
^^

ON

©

ON

ON

00

NO

-J

OO NO
4
^. O N
^^ NO LA LO O N
^ ^o ^-J Lo ON NO
* •

-J
NO

Lo

ON

NO
LA

00
NO
ON

-J

00

re
-J ••<
n &

ON
to NO
Lo Lo

- a

p

o

cr
T3

•8
© p p O p
H ^
i . to
i—*
tO ! >
LO
>-— LO
K) NO -o NO
',

i

i — i

©

O
I

i

p

p

o O

00

LO
LA

o

.

|

LA
00

O

H ^

00
ON

o o

',

1

LO

1

1

I

LA
4>-

1

1 1

©
©

©
©
© © © O © © o © O p p © © p
I L
to I . to I > 1 1 > -J
I > ©
l > 4^
I > ON
I . 4^.
I > LO
\ > LO
LO 4^
LA ©
ON
', >',4^ .LA
', .to
LO
L*J

1

1 1
ON

*. to o
-o.

p o p p p o
^o. ^
00
ON

I

00 NO
4*. Lo ON *". LA o
ON u> NO
4^ o
~o o o to

',

to -o o

o

O
LA
-O
Lo

1

©
I L
L*J

©
©
NO
00

p p p p O o
©
» > LA
',—> NO
i I

to to
I . -J
LO LO
', >LA
o
ON

LA
NO
ON

LA

ON

o

ON
-J

ON
NO

ON
LO

ON

O

LA

o
©

',—.NO
o
-o 4i

p p

O
LA

LA

O
LA

©
LA

4^ 00 4^ LO 4^ NO
4-* O to ^1 NO -O. ©

ON

O
ON
U)

-o

©

ON

O

ON

©

©

©

-J

©

©

©

ON

p
LA

N
NO
to O4;ON

O
ON
-O

©
LA
NO

4^- LA

©
ON
NO
LO

LO

•~J

©

©

I L
I L I L to
I . I . ',
LA
ON .
LA -J LA ^
LA
to 4i
LA to -o. O N

oo
u>

o O
LA
LA

©
ON
00
LA

©
LA
00
00

©
to
to
ON

©
ON
LO
LA

to

to
©

4^

©

4^
©

+- 4^ 4^
©

©

©

4i

LO

O

©

©

©

p
LA
ON

to to

K-

4^
©

LO

I L 1 i
I . ',
', ,to
to •
4^

©

©

NO

4^ 4^
©

o O

©
I L
LA

©

c
IB

©

gI -«
2. a
ON

00

» a

A ta
S 3

O
LA
LA

.244

K^

to
©

0 eg

Ito>
©

NO

4*

©
ON
LA

O o O o o
p p O p O O O © o © © © © © © © p © ©
©
H-. to Lo to to
H-»
to Lo OJ 4^ U^ 4^ to Ud ^. Lo ,
Lo ON
Lo 00 ON
Lo to , Lo u*
, LO
00 NO LA ON
to 00 OJ L»J LA -J O
LA 4^- ^1 4*
oo
to *. to
NO LA
LA LA
LO to
oo -o
to OO to 4^ NO to LO U> LO -J *> LO 4*. NO
4^ -fc- 4^ to 4^ 4v 4^ 4* 4^ 4* 4^ 4^ 4^ 4^ to
Cl
o o o O o o o o o O o © ©

o

4^
©

©
LA
LA
ON

©
^--i

©
LA
00

©

LA

©
ON

LO
LO

4*

©
© ©
to Lo Lo
LO
oo oo

*- LA

NO

to to to 4^ to
© © © © ©

•o

McBride Peak 1600m
McBride Peak 1800m
Onion Mountain 1550m
Onion Mountain 1360m
Top Lake
Wolf Creek
Grey Mountain 1150m
Keno Hill

_..
Site

Site
_, ,
Code
E3
E4
Wwl
Ww2
W
Nl
N2
Nn

No.
...
radii
25
18
31
26
31
29
22
63

No.
_
Trees
25
18
17
15
20
19
14
40
1920-2004
1923-2004
1870-2005
1811-2005
1888-2005
1928-2004
1926-2004
1890-2004

_. . _ ,
Start-End

Mean No.
years
74.7
72.8
82.6
138.8
90.7
66.9
69.5
81.8

Mean
„
Sens.
0.090
0.106
0.175
0.102
0.144
0.123
0.116
0.136
0.096
0.108
0.168
0.115
0.170
0.128
0.140
0.136

_.. _
Std. Dev.

Series
.
. ..
Intercorrelation
0.593
0.597
0.673
0.577
0.666
0.668
0.658
0.627

¥

0.207
0.210
0.182
0.343
0.453
0.248

-

...,.
AC(1)
40
40
40
40
40
40
40
40

_ ..
Spline
r

48

Note: Species are lodgepole pine (PI), interior spruce (Sx) and subalpine fir (Bl). Mean sens., Std, Dev. and AC(1) are mean sensitivity, standard
deviation and first-order autocorrelation, respectively, after standardization. Spline indicates 20- or 40-year cubic smoothing spline length used to
standardize the raw ring-widths. One core per tree was taken in June 2005 and two cores per tree were taken during the later samplings to
improve crossdating.

„ .
Species
_2_
Bl

Appendix B. Continued.

Appendix C. Pearson correlation significance of ring-width chronologies against mean
monthly temperatures (1) and precipitation (2) from May of the previous growth year to
September of the current growth year for the three species.
1, Average Monthly Temperature
Sx Previous year
Current year
Site M J J A S 0 N D
J F MA MJ J A S
S
CI
C2
El
E2
E3
E4
W
Wwl
Ww2
Nl
N2
N3
Nn

2, Monthly Total Precipitation
Current year
Sx Previous year
Site M J J A S 0 N D
J F MA M J J A S
S
CI
C2
El
E2
E3
E4
W
Wwl
Ww2
Nl
N2
N3
Nn . ;::!:.::;

PI Previous year
Site M J J A S O N D
S
CI
C2
El
E2
E3
E4
W
Wwl
Ww2
Nl
N2
N3
Nn

Current year
J F M A MJ J A S

PI Previous year
Site M J J A S 0 N D
S
CI
C2
El
E2
E3
E4
W
Wwl
Ww2
Nl
N2
N3
Nn

Current year
J F MA M J J A S

Bl Previous year
Site M J J A S 0 N D
S
CI
El
E2
E3
E4
W
Wwl
Ww2
Nl
N2
Nn

Current year
J F M A M J J A S

Bl Previous year
Site M J J A S 0 N D
S
CI
El
E2
E3
E4
W
Wwl
Ww2
Nl
N2
Nn

Current year
J F MA M J J AS

•

••

•

•

•

•

•
•

•
••

•

•

m
m

1

•

•

•

W^M

•

•

•

m

Note: Species are interior spruce (Sx), lodgepole pine (PI) and subalpine fir (Bl). Significance are
• P< 0.001,
0.001 < P < 0.01,
0.01<P<0.05

49

Chapter 3. Growth responses of three coexisting conifer species at
altitudinal treelines in the central interior of British Columbia.
Abstract
Tree-ring analyses were used to compare radial growth-climate relationships among three
coexisting, ecologically distinct conifer species at altitudinal treelines in the Engelmann
Spruce-Subalpine Fir forests in the central interior of British Columbia, Canada. Study species
included lodgepole pine {Pinus contorta var. latifolia), Engelmann spruce {Picea engelmannii)
and subalpine fir (Abies lasiocarpd). Our data showed that Engelmann spruce were positively
correlated with June and July temperatures, whereas lodgepole pine and subalpine fir were
positively correlated with October-March temperatures and Pacific Decadal Oscillation (PDO)
prior to growth. Temperature, more than precipitation, also appeared to have a stronger
influence on the tree growth. Our data suggest that the populations of the same species growing
at geographically different sites may show more similar growth-climate correlations than
different species coexisting in a community. The high values of October-March PDO are often
associated with reduced snowpack in western North America, which may extend the duration
of the growing season and consequently enhance tree radial growth at altitudinal treelines.
Winter warming may lead to increase in the abundance of lodgepole pine in the spruce-fir
forests at high-elevations in the interior of British Columbia.

50

3.1. Introduction
Trees at altitudinal treelines in the extratropical regions often experience severe environmental
conditions, including cool summer temperatures, short growing seasons, deep snowpack,
mechanical injuries due to wind-blown snow crystals, winter desiccation, low soil
temperatures, shallow soils and limited nutrients due to slow decomposition rates (Stevens and
Fox 1991, Korner 2003). Studies suggest that cold air and soil temperatures are probably the
primary factors determining altitudinal treeline positions globally (Korner 1998, 2003, Korner
and Pausen 2004). Cold temperatures often limit meristematic activities (i.e., cellular division,
cell elongations), photosynthesis and carbon allocation that may result in reduced growth rate,
reproduction and seedling establishment at high-elevations (Korner 1998, Loehle 2000a,
Jobbagy and Jackson 2000, Smith et al. 2003, Korner and Pausen 2004). Consequently
changes in temperature may affect tree productivity and vigour. Studies suggest that increases
in global annual temperatures in the future may result in changes in interactions between tree
species, reorganizations of community structures and shifts in tree species' ranges to higher
elevations (Hansen et al. 2001, Grace et al. 2002, Walther 2003).

The high-elevation Engelmann Spruce-Subalpine Fir (ESSF) forests in British Columbia (BC),
Canada, may experience significant changes in community structures under projected warming
(Hamann and Wang 2006). Predicting the potential impacts of climate change on altitudinal
treelines may require a better understanding of species-specific growth responses (Zolbrod and
Peterson 1999, Davis and Shaw 2001, Thuiller 2003). In BC three ecologically distinct conifer
species can coexist at altitudinal treelines. These include an early-successional,
shade-intolerant lodgepole pine (Pinus contorta Dougl. var. latifolia Engelm.), a
mid-successional, intermediate shade-tolerant Engelmann spruce (Picea engelmannii Parry ex
51

Engelm.), and a late-successional, shade-tolerant subalpine fir (Abies lasiocarpa [Hook.])
Nutt.) (Burns and Honkala 1990). Niche theory suggests that coexisting species should exhibit
different resource requirements and optima to coexist (Bazzaz 1987, He et al. 2005) and thus,
growth responses to climate change may differ among species. For example, Green (2005,
2007) reported that height growth responses to late growing season temperatures varied among
coexisting lodgepole pine, Engelmann spruce and subalpine fir in the central interior of BC.
Green (2005) suggests that future climate change may have more impact on the growth of
lodgepole pine than that of subalpine fir because high-elevation lodgepole pine showed the
strongest sensitivity while subalpine fir showed the weakest sensitivity to the extended
growing season conditions into fall. Coexisting species grow under the same or similar
environmental conditions and thus, they may provide insight into species-specific growth
responses to climate and underlying mechanisms.

Tree-ring analyses often provide information on the long-term impacts of climate on tree radial
growth due to the strong associations found between annual rings and climate records in the
temperate regions (Fritts 1976). Tree radial growth may be a good indicator of vigour and
competitiveness because larger radial growth is often associated with higher mechanical
stability, larger height and volume (Givnish 1995) and higher reproductive and dispersal
abilities (Despland and Houle 1997, Loehle 2000b).

Radial growth-climate relationships may be site- and species-specific because of the
considerable variations in environmental conditions (i.e., elevation, latitude) and adaptive
traits (i.e., physiology, morphology) found among species. Graumlich (1993), for example,
suggests that deciduous and coniferous trees respond to climate differently due to the different
52

size and arrangement of xylem cells, water relations and photosynthetic capacities. Villalba et
al. (1994) also reported that correlations between radial growth and monthly temperatures and
precipitation varied among lodgepole pine, Engelmann spruce and subalpine fir at mesic
high-elevation sites in the Colorado Front Range. They also found that important monthly
climate variables differed within species between mesic and xeric sites. Given the considerable
variation in topographic features, habitats, and local climate in high-elevation mountain
landscapes, the growth-climate relationships of trees may not be uniform across sites. Thus,
differences in local climate conditions may need to be considered.

Regional climate can also have significant impacts on tree growth as well as local climate
(Millar et al. 2004, Holman and Peterson 2006). The Pacific Decadal Oscillation (PDO) is a
large-scale interdecadal climate variability that influences the local climate of the Pacific
Northwest including BC (Mantua and Hare 2002, Biondi et al. 2001, MWLAP 2002). PDO,
often expressed as index series, has positive and negative phases that alternate every 20-30
years. The negative values of PDO index are generally associated with cold and wet winters
and the positive values are generally associated with warm and dry winters. Studies reported
that the radial growth of subalpine fir (Ettl and Peterson 1995, Peterson et al. 2002), lodgepole
pine (Case and Peterson 2007) and mountain hemlock (Peterson and Peterson 2001) at
high-elevations showed positive correlations with winter PDO in the Cascade and Olympic
Mountains in Washington. The potential influences of PDO on conifer growth were mainly
reported for coastal populations but rarely for interior populations in western North America.
Detecting a signal in tree growth may help clarify the existence and geographic extent of PDO
and its impacts on high-elevation ecosystems in BC.

53

In this study, we compared the growth responses and sensitivities of coexisting lodgepole pine,
Engelmann spruce and subalpine fir to climate variables at altitudinal treelines in the ESSF
zone in the central interior of BC. Growth response is defined as the significant relationship
between tree growth and climate variables, and growth sensitivity is defined as the direction
and strength of the relationships. Three hypotheses included: 1) temperature has stronger
influence on limiting tree radial growth than precipitation, 2) the growth responses of tree
populations to climate vary among species more than among sites, and 3) the large-scale
decadal oscillation, PDO, influences the radial growth of interior tree populations at altitudinal
treelines in central BC.

3.2. Methods
3.2.1. Study sites
Engelmann Spruce-Subalpine Fir (ESSF) forests occur in mountainous terrain and form
uppermost elevation forests in central British Columbia (BC) (Meidinger and Pojar 1991). The
climate of ESSF is characterized as long cold winters and short cool summers. Mean monthly
temperatures remain below 0 °C for 5 to 7 months and exceed 10 °C for 0 to 2 months each year.
Precipitation varies considerably within the zone, and 50-70 % of precipitation occurs as snow
and maximum snowpack depth ranges between 1-4 m. Three sample sites were located on the
southwest facing slopes in the moist interior subzones of ESSF (Table 3.1, Figure 3.1). Onion
Mountain is located in the lee side of the Coast Mountains where dry air from the mountains
results in relatively low annual precipitation. McBride Peak is located on the windward side of
the Canadian Rocky Mountains and receives the highest amounts of rainfall and snowfall
among the three sites (Table 3.1). The distance between Onion Mountain and McBride Peak is
approximately 600 km (Figure 3.1). Top Lake is located between the two sites and shows
54

similar climate conditions to Onion Mountain.

It is difficult to define altitudinal treelines because they often lack distinct lines in mountainous
regions. Treelines in this study were defined as transition zones between the upper limit of
continuous canopy forest and the highest patches of upright trees (Korner 1998). The sample
stands were more than 70 years old, naturally regenerated, and mid- to late-successional stages.
We selected healthy stands with the minimal visible evidence of stand level disturbance to
minimize non-climatic influences on tree radial growth. Subalpine fir and Engelmann spruce
co-dominated and lodgepole pine occurred in low abundance at the three sites.

3.2.2. Chronology development
One or two tree increment cores were extracted at breast height (1.3 m) from a minimum of
20 trees per species per site in 2005-2006. A conventional coring height of 1.3 m was chosen
because it provided sufficient growth ring records of longer than 50 years. Healthy,
canopy-dominant trees with little observable damage were selected for coring to minimize
non-climatic variation in ring-widths. Standard dendrochronology techniques were applied to
develop site- and species-specific tree-ring chronologies (Fritts 1976). Sampled cores were
mounted and sanded with increasing grain size (to 600 per inch) to observe annual rings
clearly (Stokes and Smiley 1968). Crossdating is a procedure to identify the exact year of
ring formation by synchronously matching ring-width patterns among all sampled cores from
a given site. Crossdating helps to detect any missing or false rings, because trees may fail to
produce an annual ring or may produce two rings in a year. Narrow rings were used as
pointer-years during visual crossdating (Yamaguchi 1991). Annual ring-widths were
measured to 0.01 mm using the computer program WinDENDRO™ (Regent Instruments Inc.
55

2005). The Velmex ring-measurement system (Velmex Inc. 1992) interfaced with
MeasureJ2X (VoorTech Consulting 2004) with precision 0.001 mm was used to measure
small rings. A computer program COFECHA (Grissino-Mayer 2001) was used to statistically
detect potential errors and to validate visual crossdating. Cores that did not crossdate (with a
critical threshold intercorrelation value of 0.36 based on 40-year segment with 20-year lag)
were re-examined under a microscope and remeasured using MeasureJ2X. Those cores that
did not crossdate after the re-examination were removed from the final chronology
development.

The raw ring-width measurements for each crossdated series were standardized by converting
to dimensionless ring-width indices to minimize non-climatic factors and autocorrelations
(Fritts 1976). Non-climatic factors influencing ring-width variation may include tree age,
competition, disturbance and random variation (Cook 1985). Younger trees, for instance,
generally have higher photosynthetic rates than older trees. This age-dependent biological
trend often results in the formation of wider annual rings during the early years of growth and
narrower annual rings during the later years regardless of climate condition. Changes in
interspecific and intraspecific competitions and below-ground characteristics also influence
low-frequency variation in ring-widths. Ring-width series may retain autocorrelation because
1) physiological processes within a tree often cause a lag in growth responses to climate and 2)
climate conditions tend to persist from one year to the next.

Each core was standardized by fitting a cubic smoothing spline with a 50% frequency response
cutoff of 40 years using a program ARSTAN (Cook 1985). The ring-width indices of
individually standardized cores were then averaged among all crossdated cores in a population
56

to develop a ring-width chronology for each species at each site. The Arstan chronology,
autocorrelation removed and pooled autoregression (common persistence) built back in, was
developed to maximize climate signal in the ring-width variation (Cook and Holmes 1986).

Mean sensitivity and standard deviation were calculated to evaluate the quality of standardized
chronologies. Mean sensitivity is a measure of response to year-to-year climate variation and
standard deviation indicates low- to medium-frequency variation in the ring-width
chronologies, and higher values indicate that the ring-width chronologies contain common
climate signals (Fritts 1976, Villalba et al. 1994). Mean sensitivity is calculated as the absolute
difference between adjacent ring-widths divided by the mean of the two ring-widths.

3.2.3. Data analyses
3.2.3.1. Principal component analysis
A principal component analysis (PCA) was used to identify populations that had similar
patterns in ring-width variations in the past 50 years (1953-2002) (Graumlich 1993, Peterson
and Peterson 1994). PCA is a data clustering technique that transforms a large number of
variables into a new set of variables, called principal components (PCs), which are weighted
linear combinations of the original variables. The PCs are uncorrelated with each other and the
first few PCs account for most of the variability in the original variables. The factor loadings of
ring-width chronologies are based on the correlation matrix and provide information on
similarities and differences among populations. Ring-width chronologies with positive factor
loadings are positively correlated with the resulting PC. The extracted PCs were subjected to
an orthogonal Varimax rotation to increase interpretability by maximizing the factor loadings
of loaded variables (Tabachnick and Fidell 1989). Only those PCs with eigenvalues larger than
57

1.0 were retained for further analyses. Scatter plots of the factor loadings for the PCs were used
to display the group of ring-width chronologies with similar growth patterns.

3.2.3.2. Climate model
Site-specific climate data were estimated using a climate model, ClimateBC version 3.2 (Wang
et al. 2006). ClimateBC requires latitude, longitude, and elevation to generate site-specific
monthly, seasonal and annual climate variables in western Canada based on PRISM
(parameter-elevation regressions on independent slopes model). PRISM is a regression-based
model that incorporates geographic influences on climate, including elevation, aspect, coastal
effects and orographic influences (Daly et al. 2002, Hamann and Wang 2005). For example,
elevation is a strong predictor of temperature because temperature decreases almost linearly
with altitude. Based on the topographical features, PRISM adjusts reference weather station
data to estimate mean monthly temperature and precipitation for each grid at the resolution of
approximately 4 km (Daly et al. 2002). Each reference weather station is weighted based on a
distance from a target grid cell, elevation and other topographic factors, and the value of the
target is calculated. On top of the medium resolution climate data of 4 km grid, ClimateBC
uses a high-resolution digital elevation model to build a locally and temporary scale-free
climate model for finer scale climate estimates (Wang et al. 2006). The verification was
conducted by comparing predicted and observed climate data from 191 weather stations that
had sufficient climate records included in the 1951-1980 or 1961-1990 normals (Wang et al.
2006).

3.2.3.3. Growth-climate relationships
We used a 50-year period of climate data from 1953 to 2002 in the analyses due to the
58

reliability of the climate records and the increased number of climate stations in North America
after 1950 (Wang et al. 2006). The climate variables used in the analyses included mean
monthly temperatures and monthly total precipitation generated by ClimateBC, and the
monthly Pacific Decadal Oscillation indices (PDO) obtained from the National Oceanic and
Atmospheric Administration (NOAA 2006). Seventeen months extending from May of the
previous growth year to September of the current growth year were used because this period
included two complete growing seasons (Larsen and McDonald 1995, Brooks et al. 1998).

We used response function analyses (PRECON version 5.11, Fritts 1996) to compare the
importance of temperature and precipitation on ring-width indices. The response function
transforms monthly climate variables into orthogonal eigenvectors using PCA and calculates
the multiple correlation R2, the percentage of the ring-width variance explained by the
17-months of monthly temperature and precipitation, respectively (Fritts 1976). R2 were
compared between temperature and precipitation.

Pearson simple linear correlation coefficients were calculated to identify climate variables
significantly correlated with the ring-width indices (P < 0.05). Based on the correlations, we
selected predictor climate variables, which had strong correlations and were common across
the sample populations. The direction and strength of the regression relationships (bl) between
the ring-width indices and predictor climate variables were determined using simple linear
regression analyses and then compared among species and sites using separate slopes analyses
(P < 0.05). Normality was tested using the Kolmogorov-Smirnov normality test. The analyses
were conducted using the statistical software SPSS (SPSS Inc. 1999).

59

3.3. Results
3.3.1. Chronology statistics
At each site, lodgepole pine had the highest mean sensitivity and standard deviation compared
with Engelmann spruce and subalpine fir (Table 3.2). Within species, the ring-width
chronologies at Onion Mountain and Top Lake had higher mean sensitivities and standard
deviations than those at McBride Peak. Previous studies suggest that a mean sensitivity range
of 0.095-0.166 is sufficient to make growth-climate comparison for the three species (Villalba
et al. 1994, Ettl and Peterson 1995).

3.3.2. Ring-width variation explained by temperature versus precipitation
Temperature had higher percentages in explaining the ring-width variation than precipitation
for eight out of the nine populations (Table 3.3). The simple correlation analyses also identified
stronger and more significant relationships between the ring-width chronologies and mean
monthly temperatures compared with monthly precipitation (Figure 3.2).

3.3.3. Similarity among ring-width chronologies
The first three principal components (PCs) together accounted for 76.8 % of the total variance
in the nine ring-width chronologies for the period 1953-2002 (Table 3.4). The scatter plots of
the factor loadings for the PCs showed the clustering of populations with similar growth
patterns if the loadings were greater than 0.5 (Kaiser 1974) (Figure 3.3). The first component
(PCI) showed the greatest dissimilarity in loadings between Engelmann spruce and lodgepole
pine. All pine populations had higher loadings on the PCI than subsequent components.
Subalpine fir at Onion Mountain and Top Lake also loaded on the PCI. The principal
component 2 contained spruce at Onion Mountain and Top Lake, and the principal component
60

3 contained spruce and fir at McBride Peak. The scatter plots showed that all pine populations
clustered together whereas fir and spruce at McBride Peak did not cluster with other
populations within species (Figure 3.3).

3.3.4. Growth-climate relationships
3.3.4.1. Identification of predictor climate variables
We identified four common climate variables most strongly correlated with the ring-width
indices during the 1953-2002 period based on the correlation analyses (Figure 3.2, 3.4). They
were June and July temperatures of the current growth year and October-March temperatures
and October-March PDO prior to growth (Figure 3.5). Seven out of the nine ring-width
chronologies had positive correlations with July temperatures of the current growth year. June
temperatures appeared to be also important for spruce growth at all sites. The monthly
temperatures from October to March prior to growth were averaged to form a seasonal climate
variable based on the correlations found for pine and fir (Figure 3.2). Subalpine fir showed
strong correlations with the monthly Pacific Decadal Oscillation (PDO) while Engelmann
spruce showed a few significant correlations with PDO (Figure 3.4). PDO indices from
October to March prior to growth were average and used for the further analyses because 1)
this duration corresponded to the October-March temperatures and thus, the impacts of PDO
and temperatures on tree growth can be easily associated, 2) PDO generally has the strongest
influences on the local climate of western North America from October to March (Biondi et al.
2001), and 3) several studies reported the importance of PDO for tree radial growth in the
winter months (i.e., Ettl and Peterson 1995, Gedalof and Smith 2001, Peterson et al. 2002,
Larocque and Smith 2005). October-March PDO and temperatures were also positively
correlated at each site (Table 3.5).
61

3.3.4.2. Regression and separate slopes analyses
Engelmann spruce ring-width indices had positive correlations with June and July
temperatures of the current growth year at each site (Figure 3.5). Separate slopes analyses
showed that the regression coefficients of the spruce ring-width indices to temperatures in June
(P > 0.26) and in July (P > 0.25) did not differ among sites. Lodgepole pine and subalpine fir
had positive correlations with October-March temperatures and PDO prior to growth at each
site (Figure 3.5). Separate slopes analyses showed that the regression coefficients of pine
ring-width indices to temperatures (P > 0.56) and to PDO (P > 0.053) did not differ among sites.
Subalpine fir at Onion Mountain had a higher regression coefficient than fir at McBride Peak
for temperatures (P = 0.048) and for PDO (P = 0.03). However, the regression coefficients of
subalpine fir did not differ between Onion Mountain and Top Lake (P = 0.40 for temperatures,
P = 0.65 for PDO) or Top Lake and McBride Peak (P = 0.35 for temperatures, P = 0.11 for
PDO). Subalpine fir at Onion Mountain and Top Lake were also positively correlated with July
temperatures of the current growth year, and the regression coefficients did not differ between
the two sites (P = 0.93).

3.4. Discussion
Results from this study suggest that 1) temperature probably affects tree radial growth more
than precipitation, 2) the populations of the same species growing on geographically distinct
sites may show more similar growth patterns and growth-climate correlations than among
different species coexisting in a community, 3) the Pacific Decadal Oscillation (PDO)
influences the radial growth of interior populations, but the effects are species-specific.

62

3.4.1. Importance of temperature versus precipitation
Temperature appeared to influence tree radial growth more strongly than precipitation in this
study, probably because sufficient moisture was available for tree growth in the moist ESSF
treelines. It is general that the ratio of precipitation-to-evapotranspiration increases with
altitude, leading to greater moisture surplus at high-elevation ecosystems (Korner 2003).
Adaptive strategies to moisture deficit are rare for high-elevation trees, except for the tolerance
to desiccation stress related to cold temperatures (Korner 2003). Contrary, several studies
suggest that high-elevation trees have developed adaptive strategies to cold temperatures, such
as increased photosynthesis at cool summer temperatures and freezing tolerance in winter
(Sakai and Larcher 1987, Grace et al. 2002). Thus, future changes in temperature may be more
important than precipitation in predicting tree growth at mesic treelines.

3.4.2. Similarity among ring-width chronologies
Species may be more important than site conditions in predicting tree responses to future
climate at altitudinal treelines across a large geographic range in British Columbia (BC). Our
data agreed with previous studies (Graumlich 1993, Peterson and Peterson 1994) that the
growth-climate relationships were more similar among the populations of the same species
growing at geographically different sites than different species coexisting in a site. Peterson
and Peterson (1994), for example, reported these patterns among high-elevation Engelmann
spruce, subalpine fir and subalpine larch (Larix lyallii Pari.) in the North Cascade Mountains in
Washington.

Although some species (i.e., Engelmann spruce) may exhibit a common climate signal over
large geographical areas (Makinen et al. 2000, 2002, St. George and Luckman 2001, Wilson
63

and Luckman 2003), site conditions may be important for other species in predicting growth
responses to climate. In this study, subalpine fir at McBride Peak showed different growth
responses and sensitivities to the seasonal climate conditions compared with fir populations at
Onion Mountain and Top Lake. Subalpine fir may be more sensitive to slight differences in site
conditions than Engelmann spruce or lodgepole pine, although the adaptive mechanisms
remain unknown.

3.4.3. Important growth-climate relationships
Engelmann spruce had positive correlations with June-July temperatures, which may be due to
higher day- and night-time temperatures that enhance photosynthesis and carbohydrate
allocation at treelines where growing season temperatures are generally low (Korner 1998).
Other possible causes of the positive correlations include higher photosynthetically active
radiation due to more sunny days (Goldblum and Rigg 2005), the extension of the growing
season (Danby and Hik 2007) and/or the combination of these factors. High June-July
precipitation may cause reduced spruce growth due to limited root activities in wet soils
(Kramer and Kozlowski 1979), and/or cool, cloudy summer conditions associated with high
precipitation (Table 3.5). However, high water content of soils is unlikely to cause growth
reduction in mature trees during summer at mesic sites (Korner 2003). Thus, the influence of
temperatures may be stronger than that of precipitation in limiting spruce growth in this study.

Subalpine fir and lodgepole pine had strong correlations with October-March temperatures and
PDO prior to growth in this study, suggesting that winter PDO can have significant influence
on tree radial growth at altitudinal treelines in the interior regions of BC. Several studies
reported the positive impacts of winter temperatures and PDO on the radial growth of
64

high-elevation conifers in the Coastal Mountain regions of western North America, probably
associated with the snowpack and duration of the growing season (Ettl and Peterson 1995,
Gedalof and Smith 2001, Peterson et al. 2002, Larocque and Smith 2005). The negative values
of winter PDO are often accompanied by cold temperatures and high precipitation with above
average snowpack in the Pacific Northwest including BC (Mantua and Hare 2002, MWLAP
2002). Deep snowpack generally maintains low air and soil temperatures into late spring to
early summer, which may lead to delayed bud burst, reduced shoot elongation and radial
growth (Graumlich and Brubaker 1986, Ettl and Peterson 1995). For example, Peterson and
Peterson (1994) suggest that soil temperatures and the accumulation of growing-days after
snowmelt may determine the timing of bud burst and the height growth initiation of subalpine
fir. Several tree-ring studies showed that high-elevation subalpine fir (Peterson and Peterson
1994, Ettl and Peterson 1995, Peterson et al. 2002, Larocque and Smith 2005) and lodgepole
pine (Case and Peterson 2007) had negative correlations with winter precipitation, spring
snowpack and PDO prior to growth in the Pacific Northwest extending from the western
Cascade Mountain in Oregon to the Coast Mountains in BC. In the Alps, several authors
reported stronger influences of winter temperatures than summer temperatures on the radial
growth of stone pine (P. cembra L.) at high-elevation sites where snow played important
physical roles in the ecosystems (Oberhuber 2004, Pfeifer et al. 2005, Carrer et al. 2007).
Although no monthly snowpack data were available in this study, reduced snowpack was
expected in the positive PDO years because October-March PDO had positive correlations
with temperatures and negative correlations with precipitation at most sites (Table 3.5). Actual
snow date may help in clarifying the PDO-snow relationships at each site and in examining
correlations between ring-width chronologies and snow variables such as snow depth and
timing of snowmelt in spring (Oberhuber 2004, Pfeifer et al. 2005).
65

3.4.4. Interspecific differences and adaptive mechanisms
Coexisting species experience similar external conditions including photoperiod, local climate
and soil properties and thus, interspecific growth-climate relationships may result from
species-specific internal factors such as physiological and phenological processes. For
example, studies suggest that the sensitivities of trees to cold soils differ among species
(Varpaavuori et al. 1992, Landhausser et al. 2001). Cold soils often limit root growth and water
uptake by trees due to high water viscosity and decreased root permeability, which often limit
the growth activities of aboveground tissues including photosynthesis and shoot expansion
(Day et al. 1989, Landhausser et al. 2001). However, Day et al. (1989) suggest that spruce is
insensitive to cold soil temperatures and may initiate growth under snow cover. They reported
that Engelmann spruce seedlings showed higher net photosynthesis and root growth rates than
lodgepole pine seedlings at cold soil temperatures based on a field experiment. Growth
responses of trees to spring soil temperatures may provide possible mechanisms of unique
species-specific growth responses to climate change.

Engelmann spruce showed a unique growth-climate relationship among the three study species
possibly because spruce is more adapted to cold continental climate conditions than subalpine
fir or lodgepole pine. Geologic and genetic studies suggest that spruce existed in the
unglaciated regions in Yukon and Alaska (Cwynar and Spear 1991, Landhausser et al. 2001)
while lodgepole pine and subalpine fir existed in the coastal regions of the southwestern United
States during the most recent glacial period (Wheeler and Guries 1982, Xie and Ying 1995, Ettl
and Peterson 2001). Although the three study species occupied BC after the ice retreated, their
migration routes seemed to be different. Spruce might have been subjected to extreme cold and
dry conditions while pine and fir were subjected to relatively mild conditions. Past climate
66

conditions that each species experienced may determine the adaptive responses of trees to the
current climate conditions. Biogeography of species, therefore, may provide insights into the
adaptive mechanisms of unique growth-climate relationships.

3.4.5. Implications and future research
Our data suggest that tree species may respond to future climate differently, resulting in shifts
in species dominance, abundance and distribution in the interior ESSF forests in BC. For
example, lodgepole pine and subalpine fir may become more competitive than Engelmann
spruce under warm winter conditions, resulting in increased abundance of pine and fir in
subalpine habitats. Community shifts may occur most noticeably at altitudinal treelines in
interior regions where magnitude of climate change, especially during winter time, is expected
to be large (Christensen et al. 2007). General circulation models predict that winter
temperatures will increase more than summer temperatures and more winter precipitation will
occur as rain in high-elevation regions in North America (Christensen et al. 2007). Therefore,
changes in winter conditions may need to be considered to assess the potential impacts of
climate change on forests at treelines.

Further studies on age-specific growth responses to climate may provide a better prediction of
community shifts, because younger trees may show different responses to winter warming
from mature trees. Warmer winters may favour growth and establishment of seedlings at
high-elevations due to earlier snowmelt and extended growing seasons (Smith et al. 2004).
However, warm winters may cause trees to develop weak cold hardiness and break dormancy
early in spring, which may increase the risks of cold injuries when extreme cold weather and
spring frost occur (Aitken and Hannerz 2001, Bertrand and Castonguay 2003). Trees may also
67

increase respiration and consume large amounts of carbohydrate reserves during warm winter,
which may result in lower amounts of carbohydrates for spring flush and growth (Lebourgeois
2000). Younger trees were reported to be more sensitive to extreme cold events (Van Der Kamp
and Worrall 1990) and to the loss of carbon reserves through respiration during warm winter
(Korner 1998) than mature trees. Tree developmental stages may need to be considered to
accurately predict how trees respond to winter warming.

68

3.5. Literature cited
Aitken, S.N., and Hannerz, M. 2001. Genecology and gene resource management strategies for
conifer cold hardiness. In Conifer cold hardiness. Edited by F.J. Bigras and S.J.
Colombo. Kluwer Academic Publishers, Dordrecht, Netherlands, pp 305-333.
Bazzaz, F.A. 1987. Experimental studies on the evolution of niche in successional plant
populations. In Colonization, succession and stability. Edited by A.J. Gray, M.J.
Crawley, P.J. Edwards. Blackwell Scientific, Oxford, pp 245-271.
Bertrand, A., and Castonguay, Y. 2003. Plant adaptations to overwintering stresses and
implications of climate change. Canadian Journal of Botany. 81: 1145-1152.
Biondi, F., Gershunov, A., and Cayan, D.R. 2001. North Pacific decadal climate variability
since 1661. American Meteorological Society. 14: 5-10.
Brooks, J.R., Flanagan, B., and Ehleringer, J.R. 1998. Responses of boreal conifers to climate
fluctuations: indications from tree-ring widths and carbon isotope analyses. Canadian
Journal for Forest Research. 28: 524-533.
Burns, R. M., and Honkala, B.H. 1990. Silvics of North America: 1. Conifers; 2. Hardwoods.
Agriculture Handbook 654. U.S. Department of Agriculture, Forest Service,
Washington, DC. vol.2, 877 pp.
Carrer, M., Nola, P., Louis, E., Motta, R., and Urbinati, C. 2007. Regional variability of
climate-growth relationships in Pinus cembra high elevation forests in the Alps. Journal
of Ecology. 95: 1072-1083.
Case, M.J., and Peterson, D.L. 2007. Growth-climate relationships of lodgepole pine in the
North Cascade National Park, Washington. Northwest Science. 81: 62-75.
Christensen, J.H., Hewitson, B., Busuioc, A., Chen, A., Gao, X., Held, I., Jones, R., Kolli, R.K.,
Kwon, W.-T., Laprise, R., Magana Rueda, V., Mearns, L., Menendez, C.G., Raisanen,
J., Rinke, A., Sarr A., and Whetton, P. 2007. Regional Climate Projections. In: Climate
Change 2007: The Physical Science Basis. Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and
H.L. Miller (eds.). Cambridge University Press, Cambridge, United Kingdom and New
York, NY, USA.
Cook, E.R. 1985. A time series analysis approach to tree ring standardization. Ph.D.
dissertation, University of Arizona, Tucson,
Cook, E.R., and Holmes, R.L. 1986. Users Manual for Program ARSTAN. Laboratory of
Tree-ring Research, University of Arizona, Tucson, Arizona, USA.

69

Cwynar, L.C., and Spear, R.W. 1991. Reversion of forest to tundra in the central Yukon.
Ecology. 71:202-212.
Daly, C, Gibson, W. P., Taylor, G. H., Johnson, G.L., and Pasteris, P. 2002. A knowledge-based
approach to the statistical mapping of climate. Climate Research 22: 99-113.
Danby, R.K., and Hik, D.S. 2007. Responses of white spruce (Picea glauca) to experimental
warming at a subarctic alpine treeline. Global Change Biology. 13: 437-451.
Davis, M.B., and Shaw, R.G. 2001. Range shifts and adaptive responses to Quaternary climate
change. Science. 292: 673-679.
Despland, E., and Houle, G. 1997. Climate influences on growth and reproduction of Pinus
banksiana (Pinaceae) at the limit of the species distribution in eastern North America.
American Journal of Botany 84: 928-937.
Ettl, G.J., and Peterson, D.L. 1995. Growth-response of sub-alpine fir (Abies lasiocarpa) to
climate in the Olympic Mountains, Washington, USA. Global Change Biology. 1:
213-230.
Ettl, G.J., and Peterson, D.L. 2001. Genetic variation of subalpine fir (Abies lasiocarpa (Hook.)
Nutt.) in the Olympic Mountains, WA, USA. Silvae Genetica. 50: 145-153.
Fritts, H.C. 1976. Tree rings and climate. The Blackburn Press. Caldwell, NJ. 567 pp.
Fritts, H.C. 1996. Quick help for PRECON now called PRECONK version 5.11. The
University of Arizona, Tucson Arizona.
Givnish, T.J. 1995. Plant stems: biomechanical adaptation for energy capture and influence on
species distributions. In Plant stems: physiology and functional morphology. Edited by.
G.L. Barbara. Academic Press. San Diego, California, pp 3-49.
Goldblum, D., and Rigg, L.S. 2005. Tree growth response to climate change at the
deciduous-boreat forest ecotone, Ontario, Canada. Canadian Journal of Forest
Research. 35: 2709-2718.
Grace, J., Berninger, F., and Nagy, L. 2002. Impacts of climate change on the tree line. Annals
of Botany. 90: 537-544.
Graumlich, L.J. 1993. Response of tree growth to climatic variation in the mixed conifer and
deciduous forests of the upper Great Lakes region. Canadian Journal of Forest
Research. 23: 133-143.
Graumlich, L.J., and Brubaker, L.B. 1986. Reconstruction of annual temperature (1590-1979)
for Longmire, Washington, derived from tree rings. Quaternary Research. 25: 223-234.

70

Green, D.S. 2007. Controls of growth phenology vary in seedlings of three, co-occurring
ecologically distinct northern conifers. Tree Physiology. 27: 1197-1205.
Green, D.S. 2005. Adaptive strategies in seedlings of three co-occurring, ecologically distinct
northern coniferous tree species across an elevational gradient. Canadian Journal of
Forest Research. 35: 910-917.
Grissino-Mayer, H.D. 2001. Evaluating crossdating accuracy: a manual and tutorial for the
computer program COFECHA. Tree-Ring Research. 57: 205-221.
Hamann, A., and Wang, T. 2006. Potential effects of climate change on ecosystem and tree
species distribution in British Columbia. Ecology. 87: 2773-2786.
Hamann, A., and Wang, T.L. 2005. Models of climatic normals for genecology and climate
change studies in British Columbia. Agricultural and Forest Meteorology. 128:
211-221.
Hansen, A.J., Neilson, R.P., Dale, V.H., Flather, C.H., Iverson, L.R., Currie, D.J., Shafer, S.,
Cook, R., and Bartlein, P.J. 2001. Global change in forests: responses of species,
communities, andbiomes. BioScience. 51: 765-779.
He, J-S., Zhange, Q-B., and Bazzaz, F.A. 2005. Differential drought responses between
saplings and adult trees in four co-occurring species of New England. Trees. 19:
442-450.
Homan, M.L., and Peterson, D.L. 2006. Spatial and temporal variability in forest growth in the
Olympic Mountains, Washington: sensitivity to climatic variability. Canadian Journal of
Forest Research. 36: 92-104.
Jobbagy, E.G., and Jackson, R.B. 2000. Global controls of forest line elevation in the northern
and southern hemispheres. Global Ecology and Biogeography. 9: 253-268.
Kaiser, H.F. 1974. An index of factorial simplicity. Psychometrika. 39: 31-36.
Korner, C. 1998. A re-assessment of high elevation treeline positions and their explanation.
Oecologia. 115:445-459.
Korner, C. 2003. Alpine plant life. Functional plant ecology of high mountain ecosystems. 2nd
edition. Springer-Verlag, Berlin. 344 pp.
Korner, C, and Paulsen J. 2004. A world-wide study of high altitude treeline temperatures.
Journal of Biogeography. 31: 713-732.
Kramer, P. J., and Kozlowski, T.T. 1979. Physiology of woody plants. Academic Press. N.Y.
USA. 811pp.

71

Landhausser, S.M., DesRochers, A., and Lieffers, V.J. 2001. A comparison of growth and
physiology in Picea glauca and Populus tremuloides at different soil temperature.
Canadian Journal of Forest Research. 31: 1922-1929.
Larocque, S.J., and Smith, D.J. 2005. A dendroclimatological reconstruction of climateince
AD 1700 in the Mt. Waddington area, British Columbia Coast Mountains, Canada.
Dendrochronologia. 22: 93-106.
Larsen, C.P.S., and MacDonald, G.M. 1995. Relations between tree-ring widths, climate, and
annual area burned in the boreal forest of Alberta. Canadian Journal of Forest
Research. 25: 1746-1755.
Lebourgeois, F. 2000. Climatic signals in early wood, latewood and total ring width of
Corsican pine from western France. Annals of Forest Science. 57: 155-164.
Loehle, C. 2000a. Forest ecotone response to climate change: sensitivity to temperature
response functional forms. Canadian Journal of Forest Research. 30: 1632-1645.
Loehle, C. 2000b. Strategy space and the disturbance spectrum: a life-history model for tree
species coexistence. American Naturalist. 156: 14-33.
Mantua, N.J., and Hare, S.R. 2002. The Pacific Decadal Oscillation. Journal of Oceanography.
58: 35-44.
Meidinger, D., and Pojar, J. 1991. Ecosystems of British Columbia. BC Ministry of Forests.
Victoria, BC. 330 pp.
Millar, C.I., Westfall, R.D., Delany, D.L., King, J.C., and Graumlich, L.J. 2004. Response of
subalpine conifers in the Sierra Nevada, California, U.S.A., to 20th-century warming
and decadal climate variability. Arctic, Antarctic, and Alpine Research. 36: 181-200.
MWLAP. Ministry of Water, Land and Air Protection, British Columbia. 2002. Indicators of
climate change for British Columbia, 2002. Victoria, BC. [available online]
http:www.gov.bc.ca/wlap. 48 pp.
Makinen, H., Nojd, P., Kahle, H.P., Neumann, U., Tveite, B., Mielikainen, K., Rohle, H., and
Spiecker, H. 2002. Radial growth variation of Norway spruce {Picea abies (L.) Karst.)
across latitudinal and altitudinal gradients in central and northern Europe. Forest
Ecology and Management. 171: 243-259.
Makinen, H., Nojd, P., and Mielikainen, K. 2000. Climatic signal in annual growth variation of
Norway spruce {Picea abies) along a transect from central Finland to the Arctic
timberline. Canadian Journal of Forest Research. 30: 769-777.
NOAA. 2006. Earth System Research Laboratory in National Oceanic and Atmospheric
Administration, [online] available: http://www.cdc.noaa.gov/ClimateIndices/List
72

Oberhuber, W. 2004. Influence of climate on radial growth oiPinus cembra within the alpine
timberline ecotone. Tree Physiology. 24: 291-301.
Peterson, D.W., and Peterson, D. L. 1994. Effects of climate on radial growth of subalpine
conifers in the North Cascade Mountains. Canadian Journal of Forest Research. 24:
1921-1932.
Peterson, D.W., and Peterson, D. L. 2001. Mountain hemlock growth responds to climatic
variability at annual and decadal time scales. Ecology. 82: 3330-3345.
Peterson, D.W., Peterson, D.L., and Ettl, G. J. 2002. Growth responses of subalpine fir to
climatic variability in the pacific Northwest. Canadian Journal of Forest Research. 32:
1503-1517.
Pfeifer, K., Kofler, W., and Oberhuber, W. 2005. Climate related causes of distinct radial
growth reductions in Pinus cembra during the last 200 years. Vegetation History and
Archaeobotany. 14: 211-220.
Regent Instruments Inc. 2005. WinDENDRO™. An image analysis system for tree-rings
analysis. Regent Instruments Inc. Quebec, Canada.
Sakai, A., and Larcher, W. 1987. Frost survival of plants. Reponses and adaptation to freezing
stress. Ecological Studies 62. Springer-Verlag, Berlin. 321 pp.
Smith, W.K., Germino, M.J., Hancock, T.E., and Johnson, D.M. 2003. Another perspective on
altitudinal limits of alpine timberlines. Tree Physiology. 23: 1101-1112.
SPSS Inc. 1999. SPSS. Version 15.0. Chicago, 111.
St. George, S., and Luckman, B.H. 2001. Extracting a paleotemperature record from Picea
engelmannii tree-line sites in the central Canadian Rockies. Canadian Journal of
Forest Research. 31: 457-470.
Stevens, G.C., and Fox, J.F. 1991. The causes of treeline. Annual Review of Ecology and
Systematics. 22: 177-191.
Stokes, M.A., and Smiley, T.L. 1968. An introduction to tree-ring dating. The University of
Chicago Press. 73 pp.
Yamaguchi, D.K. 1991. A simple method for crossdating increment cores from living tree.
Canadian Journal of Forest Research. 21: 414-416.
Tabachnick, B.G., and Fidell, L.S. 1989. Using multivariate statistics. 2nd Edition. Harper
Collins Publishers. NY. 746 pp.
Thuiller, W. 2003. BIOMOD- optimizing predictions of species distributions and projecting
potential future shifts under global change. Global Change Biology. 9: 1353-1362.
73

Vapaavuori, E.M., Rikala, R., and Ryyppo. 1992. Effects of root temperature on growth and
photosynthesis in conifer seedlings during shoot elongation. Tree Physiology. 10:
217-230.
Van Der Kamp, B.J., and Worrall, J. 1990. An unusual case of winter bud damage in British
Columbia interior conifers. Canadian Journal of Forest Research. 20: 1640-1647.
Velmex Inc. 1992. The Velmex "TA" System for research and non-contact measurement
analysis. Velmex Inc., Bloomfield, N.Y.
Villalba, R., Veblen, T.T., and Ogden, J. 1994. Climatic influences on the growth of subalpine
trees in the Colorado Front Range. Ecology. 75: 1450-1462.
VoorTech Consulting. 2004. MeasureJ2X. VoorTech Consulting, Holderness, N.H.
Walther, G-R. 2003. Plants in a warmer world. Perspectives in Plant Ecology, Evolution and
Systematics. 6: 169-185.
Wang, T, Hanann, A., Spittlehouse, D.L., and Aitken S.N. 2006. Development of scale-free
climate data for western Canada for use in resource management. International Journal
of Climatology. 26: 383-397.
Wheeler, N.C., and Guries, R.P. 1982. Biogeography of lodgepole pine. Canadian Journal of
Botany. 60: 1805-1814.
Wilson, R.J.S., and Luckman, B.H. 2003. Dendroclimatic reconstruction of maximum summer
temperatures from upper treeline sites in Interior British Coumbia, Canada. Holocen.
13:851-861.
Xie, C-Y, and Ying, C.C. 1995. Genetic architechture and adaptive landscape of interior
lodgepole pine (Pinus contorta spp. latifolia) in Canada. Canadian Journal of Forest
Research. 25:2010-2021.
Zolbrod, A.N., and Peterson, D.L. 1999. Response of high-elevation forests in the Olympic
Mountains to climatic change. Canadian Journal of Forest Research. 29: 1966-1978.

74

Table 3.1. Locations and climate conditions of sample sites. Climate data are the
1961-1990 normals estimated from ClimateBC (version 3.2).
Site name
Onion Mountain
Top Lake
McBride Peak

Elevation
(m)
1550
1640
1800

Latitude
(N)
54° 48'
53° 16'
53° 20'

Longitude
(W)
126° 52'
125°10'
120° 07'

MAT
(°C)
0.2
0.2
0.4

MAP
(mm)
670
710
1340

MST
(°C)
10.4
9.0
10.3

MSP
(mm)
190
240
370

GDD
(>5°C)
624
507
736

NFFD
120
111
127

PAS
(mm)
320
320
620

Note: Mean annual temperature (MAT), mean annual precipitation (MAP), mean summer
temperature (MST), mean summer precipitation (MSP), growing degree-days (GDD), the
number of frost-free days (NFFD) and precipitation as snow (PAS). Summer is June-August.

Table 3.2. Descriptions of the Arstan chronology statistics.
Site

Species

Onion Mountain

Pine
Spruce
Fir
Pine
Spruce
Fir
Pine
Spruce
Fir

Top Lake

McBride

No.
radii
24
25
31
27
30
31
20
23
18

Chronology length
(mean no. of years)
1899-2005(84.0)
1907-2005(81.4)
1870-2005 (82.6)
1870-2005 (86.6)
1843-2005 (96.9)
1888-2005 (90.7)
1904-2004 (79.6)
1920-2004 (73.0)
1923-2004(72.8)

Mean
Sensitivity
0.201
0.151
0.175
0.241
0.160
0.144
0.163
0.104
0.106

Standard
Deviation
0.226
0.152
0.168
0.256
0.160
0.170
0.194
0.099
0.108

«W

AC (1)

0.635
0.582
0.673
0.685
0.634
0.666
0.615
0.547
0.597

0.174
0.182
0.207
0.254
0.147
0.182
0.349
0.191
-

Note: First-order autocorrelation (AC (1)) and series intercorrelation (rbar)
Table 3.3. Percent variance in ring-width chronologies explained by 17 monthly
climate variables (PRECON, version 5.11). Climate variables were mean
monthly temperatures and precipitation from May of the previous growth year
to September of the current growth year for the 1953-2002 period.
Site
Onion Mountain

species
Engelmann spruce
Lodgepole pine
Subalpine fir

Temperature
0.597
0.528
0.668

Precipitation
0.495
0.292
0.269

Interaction
0.665
0.496
0.703

Top Lake

Engelmann spruce
Lodgepole pine
Subalpine fir

0.466
0.613
0.468

0.361
0.384
0.378

0.658
0.649
0.635

McBride Peak

Engelmann spruce
Lodgepole pine
Subalpine fir

0.345
0.533
0.512

0.350
0.458
0.421

0.507
0.574
0.608

Note: Interaction (%) represents the ring-width variance explained by 34
monthly variables including both temperature and precipitation.

75

Table 3.4. Varimax orthogonally rotated factor loadings for the
PC1-PC3 of the nine ring-width chronologies for the period
1953-2002.
Species
Spruce

Pine

Fir

Site
Onion Mountain
Top Lake
McBride Peak
Onion Mountain
Top Lake
McBride Peak
Onion Mountain
Top Lake
McBride Peak
Eigenvalue
% variance

Factor loadings
PCI
PC 2
0.211
0.878
0.281
0.827
-0.14
0.478
0.798
0.231
0.853
0.168
0.745
0.077
0.662
0.459
0.581
0.458
0.358
-0.085
3.0
2.2
24.4
33.0

PC 3
0.059
0.069
0.748
-0.158
0.144
0.492
0.284
0.263
0.858
1.7
19.4

Table 3.5. Correlations between predictor climate variables for the
1953-2002 period. The correlations are positive (+), negative (-) and not
significant (ns) at P < 0.05.
i-ii• .,
n, .,
Onion
Top
McBride
Climate variables
Month
.,
.
_ ,
T ,
Mountain
Lake
Peak
Temperature x Precipitation
June-July
PDO x Temperature
October-March
+
+
+
PDO x Precipitation
October-March
ns
-

Figure 3.1.Study sites in central British Columbia, Canada. Site codes are Onion
Mountain (OM), Top Lake (TP) and McBride Peak (MP).

77

(a) Temperature

(b) Precipitation
0.6

Engelmann spruce

A S O N D

J F M A M J J A S ;

S

0.4

ig

0.2

"

0.0

1 -0.2
u

I...J

J F M A M J J A

r

LI

M J J A S Q

-0.4

0.6

Lodgepole pine

Engelmann spruce

FP

Lodgepole pine

i:
u

0.0
M J J A S

A M J

J A S

| - o ,
W -0.4
0.6

Subalpine fir
2

Subalpine fir

0.4

sa 0.2

c °-°
J F M A M J

JA

J

J A S (S

L A M B

1 -0.2
u

lOnionMnt.

DTopLake

-0.4

DMcBride Peak

Figure 3.2. Pearson correlation coefficients between ring-width chronologies and mean
monthly temperature (a) and precipitation (b) from May of the previous growth year to
September of the current growth year (horizontal axes) for the period 1953-2002. Only
significant correlations (P < 0.05) are shown.

78

- I T

M

ASx
• PI
XB1

*V

0.8
0.6

g

A 0.4

o
O

0.2

T

O

X

X
o
•

T

M |

—9-—
-0.2

0
-0.2
l

'

0.2

L

r

M

x,0.4

X

M

0.6

0.8

M

A 0.8 M

0.6
§

0.4

T

0.2

x x

p.

B
o
O

O

—9-0.2

0.2

0.4

-0.2

0.6

«L

-0.4
Component 1

Figure 3.3. Similarity of ring-width variability among the nine ring-width chronologies
according to the three axes resulting from a principal component analysis with Varimax
rotation. Site codes are Onion Mountain (O), Top Lake (T), and McBride Peak (M).

79

Engelmann spruce

UL_l

•f •? «* •£ & •$ <f f fj?

J & S ^ ^ <A

Lodgepole pine

•fr -f ^ ^ cf O* ^ <? f «*"

Subalpine fir

I

f •/

^ ^ <=?* o* #•" <f -f <?^

i* ^ / " •>-** if

lOnion Mrrt.

DMcBride Peak

DTopLake

#

Figure 3.4. Pearson correlation coefficients between ring-width indices and monthly PDO
from May of the previous growth year to September of the current growth year (horizontal
axes) for the period 1953-2002. Only significant correlations (P < 0.05) are shown.

80

July Temperature

June Temperature
0.07
0.06

20
| j 0.03
|

0.02

^

0.01
0.00

1I

Spruce

Fir

Pine

Spruce

October-March Temperature

Fir

Fine

October-March PDO

0.07 j

24
0 06-

42

0.05 -

17

0.04 -§ 0.03 - •E

0 02 - -

« 0.01
Spruce

Pine

I Onion Mnt.

Spruce

• Top Lake

Fir

Pine

• McBride Peak

Figure 3.5. Linear regression coefficients (bl) between ring-width indices and selected climate
variables for the period 1953-2002 at the three study sites. The number above each bar
indicates the coefficients of determination (R2). Only significant (P < 0.05) regression
relationships are shown.

81