THE RESPONSE OF SUBALPINE FIR (Abies lasiocarpa (Hook.) Nutt)) GROWTH AND
RESILIENCE TO CLIMATE CHANGE IN NORTHERN ROCKIES, INTERIOR
BRITISH COLUMBIA, CANADA
by
Malek Haghshenas
M.Sc., University of Tehran- Iran, 2013
B.Sc., University of Tehran- Iran, 2005

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
NATURAL RESOURCE AND ENVIRONMENTAL STUDIES

UNIVERSITY OF NORTHERN BRITISH COLUMBIA

April 2021
© Malek Haghshenas, 2021

i

Abstract
Understanding forest growth and resilience under global climate change is crucial for forest
management, to maintain wood supply for future. In this research, I focused on the response and
resilience of subalpine fir (Abies lasiocarpa (Hook.) Nut)) growth to climate variables and severe
drought events at five stands located along northern slopes of the Rockies in northern BC, Canada.
Results revealed that temperature (especially summer and previous fall) was the most important
climate factor controlling growth of subalpine fir trees at study sites. The lower the latitude, the
more summer temperature negatively effected tree growth. Also, results showed that subalpine fir
trees were resilient to very dry conditions at study sites, when considering the growth recovery
period to drought was between 1 to 2 years. The lower the latitude, the faster the trees recovered
from drought. However, subalpine fir trees located at higher latitudes were more resistant to the
effects of drought.

ii

Table of Contents
Abstract .......................................................................................................................................... ii
List of Tables: ................................................................................................................................ v
List of Figures: ............................................................................................................................. vii
Acknowledgements: .................................................................................................................... xii
Chapter One: Introduction ............................................................................................................. 1
1.1 CONTEXT .......................................................................................................................... 1
1.2 BACKGROUND ……………………................................................................................ 3
1.3 RESEARCH GAPS ............................................................................................................ 6
1.4 THESIS OUT LINE AND OBJECTIVES ......................................................................... 6
1.5 REFERENCES ................................................................................................................... 8
Chapter 2: Relationship between climate and wood anatomical properties of subalpine fir (Abies
lasiocarpa) along Rocky Mountains in northern interior British Columbia, Canada ………… .12
2.1 ABSTRACT ..................................................................................................................... 12
2.2 INTRODUCTION ............................................................................................................ 13
2.2.1 Context ...................................................................................................................... 13
2.2.2 Background ............................................................................................................... 14
2.2.3 Relationships between RW, EW, LW, and Climate.................................................. 16
2.2.4 Intra-annual wood characteristics vs annual-ring width............................................ 18
2.2.5 Wood quality and need for switching to alternative species in BC........................... 19
2.2.6 Research Gaps .......................................................................................................... 20
2.2.7 Objectives ................................................................................................................. 20
2.3 METHODS ....................................................................................................................... 21
2.3.1 Site selection ............................................................................................................. 21
2.3.2 Sample Collection and Preparation............................................................................ 24
2.3.3 Fibre measurement..................................................................................................... 25
2.3.4 Preparation of cores for dendrochronological and densitometric analysis................ 27
2.3.5 Creating Master Chronologies …….......................................................................... 28
2.3.6 Comparisons of Chronologies with climate.............................................................. 29
2.3.7 Correlation and regression analysis between climate and growth parameters ......... 29
2.3.8 Regression Analysis and Comparison of Measured and Modelled Values .............. 30
2.4 RESULTS.......................................................................................................................... 31
2.4.1 Changes in climate .................................................................................................... 31
2.4.2 Correlation between climate and fibre properties...................................................... 34
2.4.3 Climate Correlations ................................................................................................. 43
2.4.3.1 Relation between climate parameters (temperature and precipitation) and RW, EW,
LW, MXD, MND, RD.……………….……………………………………….… 45
2.4.4 Regression Analysis and Comparison of Measured and Modelled Values............... 52
2.4.4.1 Relationships between fibre properties (FL, FWT, MFA and FW) and climate …52
2.4.4.2 Relationship between temperature and RW, EWD, MND, EWW, MXD, RD ..…58
2.5 DISCUSSION ................................................................................................................... 65
2.5.1 Relationship between climate, radial growth, and fibre properties at study sites...…66
2.5.2 Temporal comparison among sites based on wood fibre property changes under
impacts of microclimate variations and macroclimate thermal phases (PDO) ….... 73
2.6 CONCLUSION.................................................................................................................. 74
2.7 REFERNCES .................................................................................................................... 77
iii

Chapter 3. Legacy effects and growth resilience in northern subalpine fir stands during severe
drought……………………….………………………………………………………86
3.1 ABSTRACT ..................................................................................................................... 86
3.2 INTRODUCTION ............................................................................................................ 87
3.2.1 Research question and objectives .............................................................................90
3.2.2 Research Gap ............................................................................................................ 91
3.3 METHODS .................................................................................................................. 92
3.3.1 Site selection.............................................................................................................. 92
3.3.2 Climate Data ............................................................................................................. 94
3.3.3 Sample Collection ..................................................................................................... 94
3.3.4 Sample Preparation of 12mm diameter Cores .......................................................... 95
3.3.5 Master Chronology.................................................................................................... 96
3.3.6 Resistance, Recovery and Resilience......................................................................... 99
3.3.7 Legacy Effects…......................................................................................................101
3.3.8 Analysis and Comparison of Measured and Modelled Values.................................101
3.4 RESULTS ....................................................................................................................... 102
3.4.1 Correlation and regression analysis for growth models .......................................... 102
3.4.2 Severe drought, legacy effect, and resilience studies in subalpine fir stands.......... 107
3.5 DISCUSSION ................................................................................................................. 110
3.6 CONCLUSION................................................................................................................ 115
3.7 REFERENCES ............................................................................................................... 117
Chapter 4: Concluding Synthesis................................................................................................ 123
4.1 Key points........................................................................................................................ 123
4.2 Considerations for Future Research................................................................................. 127
4.3 Implications for The British Columbia Forestry Industry .............................................. 129
4.4 REFERENCES ............................................................................................................... 130
Appendix 1 ................................................................................................................................. 133

iv

List of Tables:
Table 2.1. Chronology characteristics and geographic information for subalpine fir (Abies
lasiocarpa) at five sampling sites for dendroclimatic study along north-facing slopes of the
Rocky Mountains in northern British Columbia (DeLong et al. 1993; Wang et al. 2016) .. 23
Table 2.2. Cutting, preparing, and labeling of 22 cores as aggregate, samples for further fibre
analysis in Canfor Pulp Eye Center ………………………………………….…………… 27
Table 2.3. Normality statistics for ring width (RW), earlywood density (EWD), earlywood width
(EWW), latewood density (LWD), latewood width (LWW), maximum density (MXD),
minimum density (MND) and ring density (RD) in Pine Le Moray, Chingee, Averil, Grizzly
Den and McBride sites including Skewness and Kurtosis values. Skewness and Kurtosis >
+1.96 or < -1.96 were not accepted as normal data. Bold values indicate non- normal values
35
Table 2.4. Normality statistics for fibre length (FL), fibre wall thickness (FWT), micro fibril
angle (MFA) and fibre width (FW) in Pine Le Moray, Chingee, Averil, Grizzly Den and
McBride sites including Shapiro-Wilk test to validate normality of sites with small fibre
sample sizes. P-value> +0.05 is considered as normal data ………...…………………… 36
Table 2.5. Pearson correlations between climate variations and fibre length (FL) and fibre wall
thickness (FWT) for Pine Le Moray, Chingee, Averil, Grizzly Den and McBride Peak
sites. Only strongest and significant correlations have been shown. Only R > 0.7 was
accepted as significant correlation. “⁎” Values are significant at 95% (P<0.05) and “⁎⁎”
shows significance at 99% (P<0.01). Oct = October …………………………….……… 36
Table 2.6. Correlations between climate variations and, micro fibril angle (MFA) and fibre width
(FW) for Pine Le Moray, Chingee, Averil, Grizzly and McBride Peak sites. Only strongest
and significant correlations (R > 0.7) are accepted as significant correlation. “⁎” Values are
significant at 95% (P<0.05) and “⁎⁎” show significance at 99% (P<0.01) ……….……… 37
Table 2.7. Master chronology statistics for study stands (Pine Le Moray, Chingee, Averil, Grizzly
Den and McBride Peak) along north-facing slopes of Rockies in northern interior British
Columbia. ring width (RW), earlywood width (EWW), earlywood density (EWD), latewood
width (LWW), latewood density (LWD), maximum density (MXD), minimum density
(MND) and ring density (RD) …………………………………………………………..… 44
Table 2.8. Relationships between circulation indices including winter (October – March) and
spring (April-June) and temperature of Winter and Spring in study stands (P, C, A, G, M).
PDO- Pacific Decadal Oscillation. All sites are Significant at 0.99 (P < 0.01) ………..…. 45
Table 2.9. Regression analysis between climate variables (precipitation and temperature) and
Earlywood width (EWW), earlywood density (EWD), minimum density (MND), ring width
(RW), latewood density (LWD) and ring density (RD) chronologies for Pine, Chingee,
Averil, Grizzly and McBride stands along north-facing slopes of the Rockies in northern BC.
R2 values only accepted if > 0.2. All values are significant at 0.99. Combined temperature
and precipitation = precipitation of May, June, July and temperature of September, October,
November, and December; Combined precipitation and temperature = precipitation of May,
v

September, October, November and December, and temperature of May and June ...…… 59
Table 3.1. Sampling site collection information and individual chronology characteristics for five
study stands along the Rock Mountains in northern BC. Biogeoclimatic variant of each site
was determined with review of site characteristics and biogeoclimatic ecosystem
classification land management handbooks (DeLong et al. 1994) ………………………... 95
Table 3.2. Master chronology statistics for five stands within central interior British Columbia.
Statistics shown include chronology length, mean sensitivity and EPS (expressed population
signal) cut off and number of trees successfully crossdated and standardized for tree-ring
width (RW), and earlywood width (EWW) ………………………………...…………….. 98
Table 3.3. Normality statistics (Skewness and Kurtosis) for ring width and earlywood width
chronologies measured at Pine Le Moray, Chingee, Averil, Grizzly and McBride sites along
the Rocky Mountain in northern BC. Data normality tests for Skewness and Kurtosis zvalues > +1.96 or < -1.96 were not accepted as normal data. Bold values indicate non-normal
values. SE= standard error ………………………………………………………………... 98
Table 3.4. Pearson correlation coefficient analysis between growth representatives (RW/EWW)
and CMD, and between Ring width (RW) and earlywood width (EWW) of subalpine fir in
all five study stands along the north-facing slopes of the Rocky Mountains. CMD= mean
climate moisture deficit, sm= Summer, sp= Spring, *= significant correlations at 95%
confidence level, **= significant correlations at 99% confidence level. Bold figures
represent r ≥ ⁺₋ 0.3 which were accepted for legacy effect study ………………………... 103
Table 3.5. Resistance, recovery, and resilience incides of Subalpine fir after sever drought and
growth index prior, during and after severe drought events (> 2-SD) in Chingee, McBride,
Pine Le Moray and Averil stands around PG and along north-facing slopes of Rocky
Mountains during 1960-2015. Gpre = growth prior to severe drought events, Gpost = Growth
after severe droughts, Gd = growth during severe drought, Rt = resistance, Rc = recovery
and Rs = resilience …………………………………………….…………………...……. 106

vi

List of Figures:
Figure 2.1. Location of sampling sites for dendroclimatic study of subalpine fir along north-facing
slopes of Rocky Mountains in northern British Columbia. From top (highest latitude) to the
bottom (lowest latitude): subalpine fir sampled in Pine Le Moray provincial park, Mount
Chingee, Mount Averil, Grizzly Den or Sugar Bowl and McBride Peak ………………… 22
Figure 2.2. Mean annual temperature for five study sites from 1940-2018. Average temperature
has generally increased at study sites over time since 1940. A= Pine Le Moray; B=Mount
Chingee; C=Mount Averil; D=Grizzly Den; E=McBride Peak. The dashed trendline
represents a 10-year moving trend line in temperature variation. Thin smooth lines represent
trendlines during 1940- 2018 period ……………………………………………………… 32
Figure 2.3. Total annual precipitation for five study sites from 1940-2018. Annual precipitation
has generally decreased at the Pine Le Moray and Chingee sites over time while increased
at McBride since 1920 (A, B, E). At the Grizzly Den and Averil sites, totals of annual
precipitation generally show steady trends since 1940 (C, D). A= Pine Le Moray; B = Mount
Chingee; C = Mount Averil; D = Grizzly Den; E = McBride Peak. A 10-year moving average
line in precipitation variation is shown with dashed line. Thin smooth lines represent
trendlines during 1940 – 2018 period. Note not all y-axis are the same ……...…………... 33
Figure 2.4. Relationships between climate variables and fibre length (FL) for Grizzly Den,
McBride Peak and Averil sites during 1920 to 2018. A: Relationship between average and
minimum temperature of December and FL (Left), and linear relation of average and
minimum temperature of December and FL at Grizzly site during 1920-2018 (Right). B:
Relationship between minimum temperature and precipitation of previous fall and FL (Left),
and linear relation of minimum temperature and precipitation of previous fall and FL at
McBride site during 1950-2018 (Right). C: Relationship between maximum temperature of
fall and summer and FL (Left), and linear relation of maximum temperature of fall and
summer and FL at Averil site during 1930- 2018 (Right). Tmean= mean temperature, Tmax=
maximum temperature, Tmin= minimum temperature, Ave= average, Pre= previous, PPT=
precipitation, STNDRD= standardized, sm= summer, FL= fibre length …………….…… 38
Figure 2.5. Relationships between climate variables and fibre wall thickness (FWT) at Pine
LeMoray, Chingee and McBride peak sites during 1940 to 2018. A: Relationship between
average and maximum temperature of May and FWT (Left), and linear relation of Average
and maximum temperature of May and FWT at Pine Le Moray during 1940-2018 (Right).
B: Relationship between maximum temperature of July and FWT (Left), and linear relation
of maximum temperature of July and FWT at Chingee during 1950- 2018 (Right). C:
Relationship between minimum temperature of previous October and FWT (Left), and linear
relation of minimum temperature of previous October and FWT at McBride during 19502018 (Right). Tmean= mean temperature, Tmax= maximum temperature, Tmin= minimum
temperature, Ave= average, Pre= previous, PPT= precipitation, STNDRD= standardized,
sm= summer, Oct= October ……………………………………………………...……….. 39
vii

Figure 2.6. Relationship between temperature and micro fibril angle (MFA) at Pine Le Moray,
Chingee and McBride Peak sites during 1940 to 2018. A: Relationship between Average
temperature of August combined with minimum temperature of July and August, and MFA
(Left), and linear relationship of Average temperature of August combined with minimum
temperature of July and August, and MFA at Pine Le Moray during 1940-2018 (Right). B:
Relationship between maximum temperature of July and MFA (Left), and linear relation of
maximum temperature of July and MFA at Chingee during 1950- 2018 (Right). C:
Relationship between minimum temperature of August and MFA (Left), and linear relation
of minimum temperature of August and MFA at McBride Peak during 1950-2018 (Right).
Tmean / ave= mean temperature, Tmax= maximum temperature, Tmin= minimum
temperature, STNDRD= standardized, sm= summer, Jul = July, Aug = August, Sept =
September …………………………………………………………..……………………. 41
Figure 2.7. Relationships between temperature and precipitation, and fibre width (FW) at Pine Le
Moray, Chingee and McBride Peak sites during 1940 to 2018. A: Relation between average
temperature of June and July combined with minimum temperature of July, and FW (Left),
and linear relation of average temperature of June and July combined with minimum
temperature of July, and FW in Pine Le Moray during 1940-2018 (Right). B: Relationship
between previous October precipitation and FW (Left), and linear relation of previous
October precipitation and FW at Chingee during 1950- 2018 (Right). C: Relationship
between fall maximum temperature and FW (Left), and linear relation of fall maximum
temperature and FW at McBride Peak during 1950-2018 (Right). Tmean / ave= mean
temperature, Tmax= maximum temperature, Tmin= minimum temperature, PPT =
precipitation, pre = previous, STNDRD= standardized, sm= summer, Oct = October, Jun =
June, Jul = July …………………………………………………………………………… 42
Figure 2.8. Correlations between mean monthly temperature and precipitation at Pine Le Moray,
Chingee, Averil, Grizzly Den and McBride Peak stands and ring width (RW), earlywood
width (EWW) along north-facing slopes of the Rockies in northern BC. Significant
correlations are depicted by ** for p < 0.01 and * for p < 0.05. Months are represented by
lower case (previous year) and upper case (current year) letters, respectively; WT =
winter,SP= spring, SM=summer, AT=autumn. S=Spearman’s correlation coefficient …... 46
Figure 2.9. Correlations between mean monthly temperature and precipitation at Pine Le Moray,
Chingee, Averil, Grizzly Den and McBride Peak stands and earlywood width (EWW),
latewood width (LWW) along north-facing slopes of the Rockies in northern BC. Significant
correlations are depicted by ** for p < 0.01 and * for P < 0.05. Months are represented by
lower case (previous year) and upper case (current year) letters, respectively; WT = winter,
SP= spring, SM= summer, AT = autumn. S= Spearman correlation coefficient …...…….. 47
Figure 2.10. Correlations between mean monthly temperature and precipitation at Pine Le Moray,
Chingee, Averil, Grizzly Den and McBride Peak stands and earlywood density (EWD) and
latewood density (LWD) along north-facing slopes of the Rockies in northern BC.
Significant correlations are depicted by ** for p < 0.01 and * for p < 0.05. Months are
represented by lower case (previous year) and upper case (current year) letters, respectively;
WT = winter, SP= spring, SM= summer, AT = autumn. S = Spearman correlation coefficient
……………………………………………………………………………………………...48
viii

Figure 2.11. Correlations between mean monthly temperature and precipitation at Pine Le Moray,
Chingee, Averil, Grizzly Den and McBride Peak stands and latewood density (LWD) and
maximum density (MXD) along north-facing slopes of the Rockies in northern BC.
Significant correlations are depicted by ** for P < 0.01 and * for P < 0.05. Months are
represented by lower case (previous year) and upper case (current year) letters, respectively;
WT = winter, SP= spring, SM= summer, AT = autumn. S = Spearman correlation coefficient
…………………………………………………………………………………………….49
Figure 2.12. Correlations between mean monthly temperature and precipitation at Pine Le Moray,
Chingee, Averil, Grizzly Den and McBride Peak stands and minimum density (MND) and
ring density (RD) along north-facing slopes of the Rockies in northern BC. Significant
correlations are depicted by ** for P < 0.01 and * = 0.05. Months are represented by lower
case (previous year) and upper case (current year) letters, respectively; WT = winter, SP=
spring, SM= summer, AT = autumn ………………………………………………….…... 50
Figure 2.13. Correlations between sum of annual precipitation at Pine Le Moray, Chingee, Averil,
Grizzly Den and McBride Peak stands and ring density along north-facing slopes of the
Rockies in northern BC. Significant correlations are depicted by ** for P < 0.01 and * for P
< 0.05. Months are represented by lower case (previous year) and upper case (current year)
letters, respectively; WT = winter, SP= spring, SM= summer, AT = autumn …...……….. 51
Figure 2.14. Measured fibre properties (solid line) vs modelled fibre properties (dotted line);
dashed lines indicate 1 decade lagged data fibre length (FL), fibre wall thickness (FWT),
micro fibril angel (MFA) and fibre width (FW) for subalpine fir at McBride Peak stand in
northern British Columbia. Data modeled from combined previous fall precipitation with
previous fall minimum temperature (A), minimum temperature of previous October (B),
September minimum temperature (C), and Fall maximum temperature (D). R2 values are
presented for ** p <0.01 and * p < 0.05. Note not all axis are the same scale ….….…….. 53
Figure 2.15. Measured fibre properties (solid line) vs modelled values (dotted line); dashed line
indicates 1 decade lagged data for fibre wall thickness (FWT), micro fibril angel (MFA) and
fibre width (FW) for subalpine fir at Pine Le Moray stand in northern British Columbia. Data
modeled from average and maximum temperature of May (A), combined average temperate
of August with minimum temperature of July and August (B), and combined average
temperature of June and July with minimum temperature of July (C). R2 values are presented
with ** p <0.01 and * p < 0.05 Note not all axis are the same scale ……………………...…55
Figure 2.16. Measured fibre properties (solid line) vs modelled values (dotted line); dashed lines
indicate 1 decade lagged data in fibre length (FL), fibre wall thickness (FWT), micro fibril
angel (MFA) and fibre width (FW) for subalpine fir at Chingee site in northern interior
British Columbia. Data modeled from July maximum temperature (A), July maximum
temperature (B), and previous October precipitation (C). R2 values are presented with ** p
<0.01 and * p < 0.05. Note not all axis are the same scale …….…………………………. 57
Figure 2.17. Measured fibre properties (solid line) vs modelled values (dotted line); dashed lines
indicate 1 decade lagged data for fibre length (FL) at Averil and Grizzly Den stands in
northern interior British Columbia. Data modeled from combined maximum temperature of
summer and fall (A), and combined December mean and minimum temperatures (B). R 2
ix

values are presented with ** p <0.01 and * p < 0.05. Note not all axis are the same scale . 58
Figure 2.18. Measured (solid line) vs modelled (dotted line) earlywood density (EWD), minimum
density (MND) and earlywood width (EWW) for subalpine fir trees at Mt. Chingee and
McBride Peak sites along the Rocky Mountains in northern British Columbia. Data modeled
from summer temperature (A), August temperature (B) and mean previous July and August
temperature (C). R2 values are presented with ** P < 0.01 and * P < 0.05. Crossed solid line
indicates 4-year lagged data. The dashed blue trendline represents a 10-year moving trend
line ……………………………………………………………...……………………….... 61
Figure 2.19. Measured (solid line) vs modelled (dotted line) ring width (RW), minimum density
(MND), ring density (RD) and latewood density (LWD) of subalpine fir trees at McBride
Peak, Grizzly Den and Averil stands along the Rocky Mountains in northern British
Columbia. Data modeled from previous July temperature (A), April temperature (B),
precipitation of May, June, July and temperature of September, October, November, and
December (C), Precipitation of May, September, October, November and December, and
temperature of May and June (D). R2 values are presented with ** P < 0.01 and * P < 0.05.
Crossed solid line indicates 4-year lagged data. The dashed blue trendline represents a 10year moving trend line ……………………………………………...…………………….. 62
Figure 2.20. Measured (solid line) vs modelled (dotted line) minimum ring density (MND) for
subalpine fir trees in Pine Le Moray site along north-facing slopes of the Rockies in northern
British Columbia during 1940 -2018. Data modeled based on combination of August
temperature and September precipitation. R2 values are presented with ** P <0.01 and * P <
0.05. Crossed solid line indicates 4-year lagged data. The dashed blue trendline represents a
15-year moving trend line ……………………………………………………………….... 63
Figure 2.21. Average winter Pacific Decadal Oscillation Index (PDO), and average winter
temperatures at Pine Le Moray, Chingee, Averil, Grizzly Den and McBride sites …….… 64
Figure 2.22. Average spring Pacific Decadal Oscillation Index (PDO), and average spring
temperatures at Pine Le Moray, Chingee, Averil, Grizzly Den and McBride sites .…….. 65
Figure 3.1. Location of sampling sites for legacy effects study in subalpine fir (Abies lasiocarpa)
forests along north-facing slopes of the Rocky Mountains in northern British Columbia.
From top (highest latitude) to the bottom (lowest latitude), subalpine fir sampled in Pine Le
Moray provincial park, Mount Chingee, Mount Averil, Grizzly Den (Sugar Bowl) and
McBride Peak ……………………………………………………………………...……... 93
Figure 3.2. Course of growth, ring width (RW) and earlywood width (EWW), in three different
stress events characterised by growth in the period before drought (PreDr), growth in the
drought period (Dr) and growth after the drought period (PostDr). Indices for resistance; Rt
= Dr ⁄ PreDr, recovery; Rc = PostDr ⁄ Dr, and resilience, Rs = PostDr ⁄ PreDr, are used to
characterize the stress response patterns. (A) Tree with growth decrease by drought (Rs=
PostDr ⁄ PreDr) if Rs< 1 indicated by low resilience (assuming drought always causes an
initial decline in growth). (B) Tree with growth recovered after drought (Rs= PostDr ⁄ PreDr)
if Rs= 1 indicated by complete resistance while Rt <1 indicated by weak resistance. (C) Tree
x

with growth considerably recovered after drought (Rs = PostDr ⁄ PreDr) if Rs> 1 indicated
by high resistance. In the graphs Rt, Rc and Rs are represented by the gradient of decline
from PostDr to Dr, the increase from Dr to PostDr, and the difference in level of PreDr and
PostDr, respectively (Presented in figure 1 by Lliot et al. 2011 and figure 1 by Pretzsch et al.
2013) …………………………………………………………………..………………... 100
Figure 3.3 Climate Moisture Deficit (CMD) during 1950-2018 in four study regions around PG
and along north facing slopes of the Rocky Mountains in northern BC. (A) CMD recognised
as drought (> 1-SD) and severe drought (>2-SD) at the Mount Chingee site where three years
(1967, 1985 and 1992) are marked as severely dry years measuring 2-SD mean CMD during
1961-2018. (B) Combined mean summer and spring CMD were used to calculate dry years
and very dry years (1960 and 1992) at the McBride stand measuring 2-SD during 1957-2018.
(C) Mean summer CMD was applied to measure dry and severely dry years (1967, 1992 and
2006) at the Pine Le Moray stand during 1956-2018. (D) Combined mean summer and spring
CMD were used to extract very dry years (1958, 1967, 1985 and 1992) during 1950-2018 at
the Averil site. Gray dotted lines represent the lowest levels at which 1-SD and 2-SD were
met. Small circle points show CMD> SD. Larger gray circles represent severe drought years
for which CMD > 2-SD. SD = standard deviation, CMD = climate moisture deficit
…………………………..……………………………………………………...……….. 105
Figure 3.4. Drought legacy during 1–4 years post severe drought years obtained from differences
between observed values and predicted values that extracted from correlations between
Ring-Width Index (RWI) or early wood width index (EWW) and climatic moisture deficit
(CMD) in four study regions around PG and along the Rocky Mountains in northern BC;
(A) legacy effect in Chingee site that longed 1 year after mean sever drought between 19602018, (B) legacy effect in McBride site that lasted 1 year after mean sever drought from 1960
to 2018, (C) legacy effect in Pine Le Moray site that longed 2 years after mean sever drought
between 1950- 2018, and (D) legacy effect in Averil stand that continued 2 years after mean
sever drought from 1940 to 2018. Dashed lines mark the level where drought happens.
=
Onset of drought event …………………….…………………………………………….. 109

xi

Acknowledgements:
I feel deeply honored and privileged to have the experience of working under the supervision of
Dr. Lisa Wood. I gratefully acknowledge her guidance and encouragement throughout the entire
course of my studies. Her supports throughout the entire research helped me overcome many
obstacles. Appreciation is also expressed to the committee members, Michael Jull, Dr. Scott Green
and Dr. Tom Pypker for providing valuable guidance during our time working together and for
their valuable comments and suggestions. Thanks to the Canfor Research staff in Burnaby for
assistance with my research analysis. Special thanks to Doug Thompson and Arsalan Mostaani for
their friendly helps all during my study at UNBC in analysing data and working with specific
software. I wish to give special thanks to my parents for their major influence on my life. Words
cannot express how appreciative I am of the love and support I have received from them. During
this journey, I met the most cherished one, the love of my life, Fatemeh. I feel fortunate to have
her unwavering support, and deeply thank her for her love, support and sacrifices and for giving
me the strength and courage to sail through challenges. Thank you for keeping life joyful for me
with your love, liveliness and beauty. Special thanks to my brother, Dr. Meysam Haghshenas for
illuminating my future path and his kindly supports.
Support and funding for this research were provided by UNBC Research Award and through a
collaborative Mitacs grant with Canfor Forest Products Ltd.

xii

Chapter 1. General introduction
Introduction
1.1 Context
Under all emission scenarios explored by the Intergovernmental Panel on Climate Change, global
average temperature and sea levels are predicted to rise. Global average temperature is projected
to rise by 1.4 °C to 5.8 °C between 1990 to 2100 (IPCC 2013), and the average global temperature
has already increased by 0.82 ºC during the period between 1880 and 2012 (IPCC 2014). Summers
are generally predicted to be warmer and drier, while winters are likely to be warmer and wetter
(Mote 2003). Climate vulnerability, in areas susceptible to disproportionately large changes in the
climate system, is projected to be particularly high (Lindner et al. 2010). In this context, cold
temperature ecosystems consisting of cold-adapted species are particularly sensitive to recent
global warming (Körner 1999). Of all forests around the world, Canada is ranked the third largest
in forested area (Canadian Forest Service 2017), and several studies have indicated that forest
habitats may be at risk because of global warming (Seager et al. 2007; Candel-Pérez el al. 2012).
The impact of climate change is evident as reflected by the 2ºC increase in average annual
temperatures in the boreal regions in Western Canada during the 1950-2003 period (Price et al.
2013). Forests are among the most sensitive ecosystems in the world, affected by global climate
change. Boreal forests in British Columbia (BC) are among vulnerable ecosystems which have
also been impacted by global climate change (Hamann and Wang 2006; Lo et al. 2010) as well as
forest ecosystems of central British Columbia (Zhang et al. 1999; Wang et al. 2012). Species at
the edge of their natural range or growing beyond their niche are considered to be particularly
vulnerable (Handewinkel et al, 2013; Seidl et al, 2011). High-altitude forests will be specifically
vulnerable if the frequency of disturbances increases considerably (Keane et al 2018). Generally,
1

temperature is considered more important to growth of tree species than precipitation in northern
interior BC (Splechtna et al. 2000; Wood and Smith 2015; Cortini et al. 2016) most probably
because this region has long and cold winters (Wood and Smith 2015). Studies in the boreal forest
of northwestern Canada and in the Canadian northern Rockies show that the annual-rings of these
species are mostly under control of summer temperature (D'Arrigo et al. 1992; Szeicz and
MacDonald 1995; Luckman et al. 1997). Adams (2014) points out that global warming is expected
to initiate long-term changes in cellular wood structure and radial growth resulting in basic shifts
in productivity and wood quality over vast areas. Consequently, BC forests may be in danger of
tracheid degradation and/or lower wood product quality for final products like pulp and paper
(Wang et al. 2012).
Fibre length is defined as the length of wood cells along the longitudinal axis, also called tracheid
length in softwoods (Van Leeuwen et al. 2011). Fibre length is an important wood quality attribute
for timber items considering that longer fibres result in higher wood quality for the pulp and paper
industry because of the positive relationship between fibre length and paper strength (Jozsa and
Middleton 1994). Furthermore, Jozsa and Middleton (1994) found that trees generally produce
large diameter, thin-walled fibres during the early part of the growing season, and smaller
diameter, thick-walled fibres in the late summer and end of growing season, both of which are also
fibre characteristics used in wood quality determination. Recently, expanded timber harvesting of
spruce (Picea sp.) alongside different species like Pinus sp., has diminished spruce supplies in
some lower-elevation habitats in the northern Rockies (Kean et al. 2018). Furthermore, the
convergence of invasive pests including mountain pine beetle (Dendroctonus ponderosae), spruce
beetle (Dendroctonus rufipennis), and expanded rapidly spreading wildfires which are for the most
part climatically driven (Abatzoglou and Williams 2016) recently in interior BC. Therefore, the
2

wood supplies tied with high wood quality needed by the wood factories are affected, especially
for future timber supply and silvicultural plans. Thus, the forest and wood products industries are
being forced to consider higher elevation forested zones and other species, such as subalpine fir
(Abies lasicarpa) for timber supplies.
1.2 Background
Subalpine fir is a broadly dispersed North American fir which territory range stretches out from
32° 52ʹ N latitude in Arizona and New Mexico to 64° 30ʹ N in the Yukon Region, Canada
(Alexander 1990). This species is one of the fundamental timber yield species in the boreal district
of British Columbia (Splechtna et al. 2000), of significance both economically and
environmentally. Subalpine fir is a shade tolerant species and in the lower elevations of the
Engelmann Spruce – Subalpine Fir (ESSF) biogeoclimatic classification zone; it is the most
abundant in the understory. However, at high elevations of this zone and in some wetter areas,
subalpine fir frequently dominates the forest canopy (Coupe et al. 1991). In the Rocky Mountains
of British Columbia and Alberta south of the Peace River, subalpine fir grows between 914 to
2134 m (3,000 to 7,000 ft), but it is more abundant above 1524 m (5,000 ft) (Alexander et al.
1984). From ecological aspects, subalpine fir habitat is restricted to cold, humid habitats. Subalpine
fir can be sensitive to some types of climate change, for example, it has low tolerance to high
temperatures (Alexander et al. 1990).
Subalpine fir can be a climate sensitive species (Zhang et al. 1999). Due to accumulation of
snowpack in subalpine fir stands, growth of this species is limited to a short growing season
(Peterson et al. 2002). Zhang (1999) showed that growth of subalpine fir is limited by June
temperature and July precipitation. However, fall temperature positively correlated with subalpine
fir growth in northwestern US (Peterson et al. 2002). Generally, at higher latitudes and on northern
3

slopes where soil is wetter, subalpine fir growth is limited by late snowpack melting and as a result
shorter growing season because of delay in start of growing season (Peterson et al 2002). At lower
latitudes and warmer sites, summer temperature could be more detrimental to growth of subalpine
fir (Ettl and Peterson 1995).
Some wood anatomical traits, including density and microfibril angle, are among the most
important determinants for wood stiffness and strength and therefore, wood quality (Zhang et al.
2020). Microfibril angle (MFA) is a measurement attributed to the angle of cellulose microfibrils
located in the secondary cell wall of fibres to the tracheid axis (Barnett and Bonham 2004). MFA
affects strength and elasticity of wood (e.g. shrinkage behavior), which influences wood quality
(Huang et al. 2003; Donaldson 2008). The decrease in microfibril angle from the pith outwards to
the bark produces higher strength mature wood in comparison to juvenile wood (Mansfield et al.
2009). Mature wood compared with juvenile wood has smaller microfibril angle and higher wood
density (Zhang et al. 2020). Therefore, wood quality can be related to rate of tree growth and
proportion of mature and juvenile wood within a stem. The higher the density, the higher the
quality of wood (Zhang et al. 2020) and strength (machinability) (Josza and Middleton 1994).
Some wood anatomical traits including proportion of earlywood and latewood, and ring width are
mostly controlled by environmental factors rather than genetic traits (Lenz et al. 2010). Wood
density can be calculated through the ratio of latewood proportion to earlywood proportion (Van
Leeuwen et al. 2011); Juvenile wood and earlywood generally have lower density values
(Mansfield et al. 2009). Radial growth rate which influences on wood quality (Zhang et al. 2020)
is both genetic and climatically controlled of which environmental parameters generally have been
shown to be more important than genetic parameters (Downes and Drew 2008). Latewood contains
dense cells attributed to the slower growth periods in the end of growing season (late summer).
4

Conversely, earlywood includes less dense cells which form in the periods of faster growth in first
months of the growing season (early summer). Generally, it can be inferred that earlywood cells
form when climate parameters including the most influential ones (temperature and precipitation)
are optimum (first months of growing seasons) and latewood cells forms when reduction starts in
amount of precipitation and temperature at the end of growing season (Sun et al. 2016). Looking
deeper into wood properties may clarify intra-annual wood property variation, which forms within
months of the year that are probably more sensitive to climate variations than tree-ring width,
which forms annually.
The resilience of forest ecosystems has recently received increased attention in response to
growing concerns about climate change, and has been proposed as an important factor in
addressing future uncertainties in ecosystem management (Biggs et al. 2012; Seidl 2014).
Meanwhile, wide-ranging changes in climate and disturbance regimes, including prolonged
drought and increased wildfire size and frequency, have raised concerns about forest resilience to
environmental climatic change (Abatzoglou and Williams 2016; Seidle et al. 2016).
Flexibility or resilience can be extensively described as a system's capacity to deal with
perturbance and to endure in the face of it (Carpenter et al. 2001) or potential ability of an
ecosystem to be able to return to the preconditions including physiological condition, composition
and structures, after disturbance (Holling 1973). Hence, a forest’s resistance to climatic stressors
depends on the trees’ ability to restore and survive (Hanklin et al. 2018). It should be considered
that stand level variables may have stronger influences on forest resilience than climate. For
instance, Seidl et al. (2017) showed that the role of forest age in forest resilience was highlighted
where old forests and overstock conditions have especially low resilience to perturbations.
Furthermore, in those stands that trees have already experienced several drought events in recent
5

past years, resilience to future disturbances may decreases (Gutschick and BassiriRad 2003; Seidl
et al. 2017). It usually takes a few years (2 to 4 years on average) for the trees to recover after
perturbance even when climate conditions return to the normal (Anderegg et al. 2015).
1.3 Research gap
Few studies have focused on the relationships between growth and climate variations in subalpine
fir in northern BC. Furthermore, most previous studies have focused on tree-ring width, and the
number of articles focusing on anatomical traits including fibre/cell wall thickness (FWT), ring
density (RD), fibre width (FW), and MFA of subalpine fir is small. Also, there is a lack of
information on the relationship between climate and forest tree resilience and study of legacy
effects in subalpine fir forests of northern interior BC, as these kinds of studies are fairly new and
were only initiated in recent years.
1.4 Research Questions and Objectives
1- How do climate variations (temperature and precipitation) affect wood anatomy (RW, EW,
LW, MXD, MND, RD, FL, MFA, FWT, FW) of subalpine fir at five geographically
different stands in northern interior BC?
Objectives:


Determine if there are any statistically significant relationships between climate
variations and radial growth of subalpine fir in northern British Columbia, and, if so,
determine and compare variation in subalpine fir’s wood properties with temperature
and precipitation at study sites.



Evaluate whether intra-annual wood characteristics show stronger relationships with
climate than do annual-ring widths, through correlation analysis.

2- Do relationships between radial growth and climate show evidence of resilience to drought,
6

in subalpine fir stands in interior BC? If so, do these relationships show differences in
drought resilience with changes in latitude in northern interior British Columbia?
Objectives:


Evaluate patterns in deviations from the average in wood measurements as indicators of
resistance, resilience or recovery after severe climate (e.g., drought) in subalpine fir forests.
Compare measurements and trends at five study sites with different latitudes.



Evaluate how long (years) it takes to subalpine fir recover after severe drought (legacy
effects) in the five study sites of the Rocky Mountains, British Columbia, Canada.

This thesis contains two data chapters that address the above questions, and a concluding chapter
in which data chapters are synthesized. Chapter 2 explores wood properties including RW, EW,
LW, RD, MXD, MND and FL, FWT, FW, MFA at five natural and unmanaged stands near Prince
George along high elevations of the Rocky Mountains in BC, Canada in relation to climate.
Chapter 3 examines resilience of subalpine fir (study of legacy effects) at five study sites with
severe drought and identifies recovery durations after severe drought in subalpine fir.
In this research, I chose to focus on latitude for this particular study, while keeping elevation and
aspect consistent. I acknowledge that significant variation also exists in tree growth based on
elevation and different aspects as well, but these were outside of the scope of this study.

7

1.5. References:
Abatzoglou JT and Williams AP. 2016. Impact of anthropogenic climate change on wildfire across western
US forests. Proceedings of the National Academy of Sciences of the United States of America, 113:1177011775.
Adams HR, Barnard HR, Loomis AK. 2014. Topography alters tree growth–climate relationships in a semi‐
arid forested catchment. Ecosphere, 5(11):1-6.
Alexander RR, Shearer RC and Shepperd WD. 1984. Silvical characteristics of subalpine fir. General
Technical Report RM-115. USDA Forest Service, Rocky Mountain Forest and Range Experiment Station,
Fort Collins, CO. RM-115: 60-70
Alexander RR and Shepperd WD. 1990. Picea engelmannii Parry ex Engelm. Engelmann spruce. Silvics
of North America, 1:187-203.
Anderegg WR, Schwalm C, Biondi F, Camarero JJ, Koch G, Litvak M, Ogle K, Shaw JD, Shevliakova E,
Williams AP, Wolf A. 2015. Pervasive drought legacies in forest ecosystems and their implications for
carbon cycle models. Science, 349(6247):528-32.
Barnett JR, Bonham VA .2004. Cellulose microfibril angle in the cell wall of wood fibres. Biological
Reviews, 79: 461-472 - doi: 10.1017/s146 4793103006377
Biggs R, Schlüter M, Biggs D, Bohensky EL, BurnSilver S, Cundill G, Dakos V, Daw TM, Evans LS,
Kotschy K, Leitch AM, et al. 2012. Toward principles for enhancing the resilience of ecosystem
services. Annual Review of Environment and Resources. 37:421-448.
Carpenter S, Walker B, Anderies JM and Abel N. 2001. From metaphor to measurement: resilience of what
to what? Ecosystems, 4(8): 765-781.
Cortini F, MacIsaac DA, Comeau PG. 2016. White spruce growth and wood properties over multiple time
periods in relation to current tree and stand attributes. Forests, 7(3) - doi.org/10.3390/f7030049
Coupé R, Stewart AC and Wikeem BM. 1991. Engelmann spruce–subalpine fir zone. Ecosystems of British
Columbia, pp.223-236.
D'Arrigo RD, Jacoby GC, Free RM. 1992. Tree-ring width and maximum latewood density at the North
American tree line: parameters of climatic change. Canadian Journal of Forest Research, 22(9):1290-1296.
Donaldson L .2008. Microfibril angle: measurement, variation and relationships - a review. Iawa Journal,
29: 345-386 - doi: 10.1163/229419 32-90000192
Downes GM and Drew DM .2008. Climate and growth influences on wood formation and utilization.
Southern Forests: a Journal of Forest Science, 70: 155-167 - doi: 10.2989/south.for.20 08.70.2.11.539
Ettl GJ and Peterson DL. 1995. Growth response of subalpine fir (Abies lasiocarpa) to climate in the
Olympic Mountains, Washington, USA. Global Change Biology, 1(3): 213-230.
8

Forest Service Publications. The State of Canada's Forests. Annual Report 2017. 2017. Natural Resources.
Hamann A, Wang T. 2006. Potential effects of climate change on ecosystem and tree species distribution
in British Columbia. Ecology, 87(11): 2773-86.
Hanewinkel M, Cullmann DA, Schelhaas M-J, Nabuurs G-J, Zimmermann NE. 2013. Climate change may
cause severe loss in the economic value of European forest land. Nature Climate Change, 3:203-207.
Hankin, L. 2018. Annual climate impacts on tree growth and post-fire regeneration in Ponderosa Pine and
Douglas fir in the northern Rocky Mountains.
Holling CS. 1973. Resilience and stability of ecosystems. Annual. Review of Ecology, Evolution and
Systematics, 4: 1-23.
Huang CL, Lindstrom H, Nakada R, Ralston J. 2003. Cell wall structure and wood properties determined
by acoustics-a selective review. European journal of wood and wood products, 61: 321-335 - doi:
10.1007/s00107-003-0398-1
IPcc. 2013. Stocker TF, Qin D, Plattner GK, Tignor M, Allen SK, Boschung J, Nauels A, XiA Y, Bex V
and Midgley PM. Climate Change .2013. The physical science basis. Contribution of working group I to
the fifth assessment report of the intergovernmental panel on climate change, Cambridge/ New York:
Cambridge University Press.
IPcc. 2014. Climate Change 2014. Impacts, adaptation, and vulnerability. Contribution of working group
II to the fifth assessment report of the Intergovernmental Panel on Climate Change. (Field CB, Barros VR,
Dokken DJ, Mach KJ, Mastrandrea MD, Bilir TE, M. Chatterjee, K. L. Ebi, Y. O. Estrada, R. C. Genova,
B. Girma, E. S. Kissel, L. A. N, M. S, P. R. Mastrandrea, and L. L. White, Eds.). Cambridge University
Press, Cambridge, United Kingdom and New York, NY, USA.
Jozsa LA, Middleton GR .1994. A discussion of wood quality attributes and their practical implications.
Special Publication SP-34, Canada, British Columbia Partnership Agreement on Forest Resource
Development, FRDA II, Forintek Canada Corporation, Vancouver, BC, Canada, pp. 42. URL:
http://pdfs.seman ticscholar.org/57b5/7386b0f5029e60812695c70 ac34eb624069b.pdf
Keane RE, Mahalovich MF, Bollenbacher BL, Manning ME, Loehman RA, Jain TB, Holsinger LM, Larson
AJ. 2018. Effects of climate change on forest vegetation in the Northern Rockies. In Climate Change and
Rocky Mountain Ecosystems: 59-95. Springer.
Körner CH. 1999. Alpine plant life. Springer-Verlag, Berlin.
Lenz P, Cloutier A, MacKay J, Beaulieu J. 2010. Genetic control of wood properties in Picea glauca: an
analysis of trends with cambial age. Canadian Journal of Forest Research, 40: 703-715. - doi: 10.1139/X10014
Lindner M, Maroschek M, Netherer S, Kremer A, Barbati A, Garcia-Gonzalo J, Seidl R, Delzon S, Corona
P, Kolström M, Lexer MJ. 2010. Climate change impacts, adaptive capacity, and vulnerability of European
forest ecosystems. Forest ecology and management, 259(4):698-709.
9

Lo YH, Blanco JA, Seely B, Welham C, Kimmins JP. 2010. Relationships between climate and tree radial
growth in interior British Columbia, Canada. Forest Ecology and Management, 259(5): 932–942 doi.org/10.1016/j.foreco.2009.11.033.
Luckman BH, Briffa KR, Jones PD, Schweingruber FH. 1997. Tree-ring based reconstruction of summer
temperatures at the Columbia Icefield, Alberta, Canada, AD 1073-1983, The Holocene, 7: 375-389.
Mansfield SD, Parish R, Di Lucca CM, Goudie J, Kang K-Y, Ott P. 2009. Revisiting the transition between
juvenile and mature wood: a comparison of fibre length, microfibril angle and relative wood density in
lodgepole pine. Holzforschung 63 - doi: 10.1515/hf.2009.069.
Mote PW. 2003. Trends in temperature and precipitation in the Pacific Northwest during the twentieth
century.
Price DT, Alfaro RI, Brown KJ, Flannigan MD, Fleming RA, Hogg EH, Girardin MP, Lakusta T, Johnston
M, McKenney DW, Pedlar JH. 2013. Anticipating the consequences of climate change for Canada’s boreal
forest ecosystems. Environmental Reviews, 21(4):322-65.
Peterson DW, Peterson DL, and Ettl GJ. 2002. Growth responses of subalpine fir to climatic variability in
the Pacific Northwest. Canadian Journal of Forest Research. 32(9), pp.1503-1517.
Seager R, Ting M, Held I, Kushnir Y, Lu J, Vecchi G, Huang HP, Harnik N, Leetmaa A, Lau NC, Li C.
2007. Model projections of an imminent transition to a more arid climate in southwestern North America.
Science, 316(5828):1181-1184.
Seidl R, Donato DC, Raffa KF, Turner MG. 2016. Spatial variability in tree regeneration after wildfire
delays and dampens future bark beetle outbreaks. Proceedings of the National Academy of Sciences.
113(46):13075-13080
Seidl R, Rammer W, Lexer MJ. 2011. Climate change vulnerability of sustainable forest management in
the Eastern Alps. Climate Change, 106:225-254.
Seidl R, Rammer W, Spies TA. 2014. Disturbance legacies increase the resilience of forest ecosystem
structure, composition, and functioning. Ecological Applications, 24:2063-2077.
Seidl R, Vigl F, Rössler G, Neumann M, Rammer W. 2017. Assessing the resilience of Norway spruce
forests through a model-based reanalysis of thinning trials. Forest ecology and management, 388:3-12.
Splechtna BE, Dobrys J, 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(2):89-100.
Szeicz JM, MacDonald GM. 1995. Dendroclimatic reconstruction of summer temperatures in northwestern
Canada since A.D. 1638 based on age-dependent modeling. Quaternary Research, 44, 256-266.
Sun Y, Wang LL, Yin H. 2016. Influence of climatic factors on tree-ring maximum latewood density of
Picea schrenkiana in Xinjiang, China. Frontiers of Earth Science 10: 126-134 - doi: 10.1007/s11707-0150507-6
10

Van Leeuwen M, Hilker T, Coops NC, Frazer G, Wulder MA, Newnham GJ, Culvenor DS. 2011.
Assessment of standing wood and fibre quality using ground and airborne laser scanning: a review. Forest
Ecology and Management, 261: 1467-1478 - doi: 10.1016/j.foreco.2011.01.032
Wood LJ, Smith DJ. 2015. Intra-annual dendroclimatic reconstruction for northern British Columbia,
Canada, using wood properties. Trees, 29(2): 461-74.
Zhang Q, Alfaro RI, Hebda RJ. 1999. Dendroecological studies of tree growth, climate and spruce beetle
outbreaks in Central British Columbia, Canada. Forest Ecology and Management, 121(3):215-25.
Zhang S, Belien E, Ren H, Rossi S, Huang JG. 2020. Wood anatomy of boreal species in a warming world:
a review. iForest - Biogeosciences and Forestry, 13(2): 130.
Zhang Q, Alfaro RI, and Hebda RJ. 1999. Dendroecological studies of tree growth, climate and spruce
beetle outbreaks in Central British Columbia, Canada. Forest Ecology and Management. 121(3): 215-225.

11

Chapter 2: Relationship between climate and wood anatomical properties of subalpine fir
(Abies lasiocarpa (Hook.) Nutt)) located in the Rocky Mountains of northern interior
British Columbia, Canada

2.1 Abstract
Understanding the response of boreal forests to climate change is critical for proper forest
management and for sustainable wood supply. In this study I focused on climate change impacts
on wood anatomy (ring width, earlywood, latewood, ring density and fibre properties) of subalpine
fir (Abies lasiocarpa (Hook.) Nutt.)) at five forest stands located along the northern aspects of the
Rocky Mountains in northern BC, Canada. Results revealed that intra-annual wood characteristics
showed stronger statistical correlations with climate indices than did simple tree-ring width orders.
Growth of subalpine fir, measured by ring widths, densities, fibre lengths, and fibre widths was
influenced largely by summer temperature, previous fall temperature, and to a less extent
September precipitation. The lower the latitude, the greater the negative effect of summer
temperatures on radial growth at study stands. Generally, temperature was the most important
climatic factor in controlling growth of subalpine fir at study sites.

12

2.2 INTRODUCTION
2.2.1 Context
Global average temperature is projected to rise by 1.4 °C to 5.8 °C between 1990 to 2100 (IPCC
2013). Meanwhile, northern latitude ecosystems seem to be more sensitive to global climate
change that may have dramatic consequences in boreal ecosystems (IPCC 2013). Over the past
decades, British Columbia, Canada (49°-60° north latitude) has experienced a warming trend that
approximately matches climate change predictions from general circulation models discussed in
the mid 1990s (Johns et al. 1997, Mote 2003). Although same general models predict climate to
be warmer and wetter in pacific Northwest of north America (Lo et al. 2010; Fleming and
Whitfield 2010), and forests of boreal regions are expected to benefit from global climate change
by enhancing their primary and secondary growth by 2100 (Moring el al. 2009), regions of British
Columbia may experience warmer and drier years along with increased drought stresses (Lo et al.
2010; Jiang et al. 2016). A growing number of studies suggests that global climate warming affects
a wide range of species and ecosystems (Walther et al. 2002, Parmesan and Yohe 2003, Root et
al. 2003).
One of the characteristics of determining the quality of wood is the density of the wood. Boreal
forests, which compose 25% of all forests on earth, yield almost 33% of lumber products and 25%
of pulp and paper exports (Burton et al. 2010). Generally, the mature wood portion of a tree’s
profile is made with dense, longer fibres or tracheids fibre length (FL), thicker fibre walls (FWT)
and lower micro fibril angel (MFA) when compared to juvenile wood (Zhang et al. 2020). These
characteristics are usually associated with higher wood quality (Plomion et al. 2001; Huang et al.
2003). Wood anatomy is controlled by both environmental and genetic traits (Downes and Drew
2008; Downes et al. 2009). Past studies demonstrate that wood anatomical traits are affected by
13

changing climate (Wood and Smith 2012; Pitman and Smith 2013; Watson and Luckman 2016).
Furthermore, some traits like earlywood width (EWW) or (EW) and latewood width (LWW) or
(LW) are shown to be more affected by environmental and climatic factors than genetic factors
(Lenz et al. 2010). Therefore, understanding the impacts of climate change on wood anatomical
traits including annual-ring-with (RW), earlywood width (EWW), and latewood width (LWW),
earlywood density (EWD), latewood density (LWD), maximum density (MXD), minimum density
(MND), and ring density (RD) is crucial to understand tree stability in changing climates. EW is
among those anatomical parameters that are directly controlled by climate variations (Nabeshima
et al. 2015)
2.2.2 Background:
EW and LW are defined as the portion of the ring in which wood density is respectively below or
above the mean wood density of the ring (Drew et al. 2013). EW forms during the first months of
the growing season when temperature and precipitation are optimal. In warmer years, the
proportion of EW in a tree-ring is higher than LW, especially at higher elevations (Wang et al.
2002; Deslauriers et al. 2003), and on south-facing, warmer slopes (Huang et al.2011). Generally,
production of large, thin-walled EW cells occurs at the onset of the growing period (Barnett and
Jeronimidis 2003). The relative production of EW can be attributed to the previous growing
season’s energy reserves. Optimal EW production occurs in June and July when photoperiod is at
highest points (Rossi et al. 2006). Conversely, LW forms at the end of the growing season when
precipitation and temperature decline. Onset and termination of LW formation in boreal forests
may range from the end of June to mid-August, depending on both species and location (Huang et
al. 2011; Zhai et al. 2012). Furthermore, that when summer is cold, formation of LW starts earlier,
and as a result the proportion of LW to EW in tree-ring width is higher (Huang et al. 2011). Huang
14

et al (2011) also found that on north-facing slopes, the proportion and formation of LW was higher
than, on south-facing slopes, likely due to colder temperatures characteristic of north-facing slopes.
LW is very sensitive to temperature and water availability since its formation is contingent upon
carbohydrate accumulation produced by photosynthesis (Sun et al. 2016). Several studies report a
negative relationship between RW and RD at higher elevations (Lindström et al. 1996; Dutilleul
et al. 1998) which may strengthen the idea of formation of denser wood at slower-grown higher
elevations.
Wood density is defined as a wood substance for a given volume of wood (Zobel and Van
Buijtenen 1989). Together, wood density, tracheid size, and wall thickness, are the most important
wood characteristics in determining wood quality. Some indicators of wood quality such as
stiffness, machinability and strength are directly determined by wood density (Wimmer et al.
2002b). Wood density changes within a tree from juvenile to mature, and from EW to LW, and
generally, lower wood density is found in juvenile and EW (Mansfield et al. 2009). It can be
inferred that juvenile trees with fast growing traits produce larger volumes of wood and
consequently lower wood density. However, mature trees and species grown in dry or cold climate
situations have slower growth rates and produce denser wood (Camarero and Gutierrez 2017).
There are some studies worked on cell size and cell wall thickness under impact of climate change
(Larson et al. 2001; Rossi et al. 2008; Rossi et al. 2011; Rossi et al. 2014). The wood production
or xylogenesis process begins in early spring with the formation of the first row of xylem cells,
which these cells have thin walls and an increase in their number and cell division usually
continues until mid-summer (Rossi et al. 2014). Later in summer and near end of the growing
season, the number of produced cells no longer changes, but cell wall thickening increase until cell
become mature and then the process of lignification has been finished (Rossi et al. 2014).
15

Therefore, cells formed in the early part of the growing season display larger radial diameter,
thinner cell walls with a larger lumen and relatively lower density (earlywood) when compared to
cells formed later in the growing season which have a smaller radial diameter, thicker cell walls
with a small lumen and higher density (Panshin and de Zeeuw 1980).
Fibre length (FL), known as tracheid length in conifers, is defined as the length of wood cells along
the longitudinal axis (Van Leeuwen et al. 2011). The longer the fibre, the better the quality of
wood and the stronger the paper as a final product (Jozsa and Middleton 1994). Climate variations,
especially temperature, affect FL (Zhang et al. 2020). Generally, in conifers, tracheids are
responsible for water transportation within the plant (Badel et al. 2015), as well as providing
mechanical support (Vaganov et al. 2006). Tracheid size is positively related with water
availability when cells are forming (Vieira et al. 2009).
Micro fibril angle (MFA) is angle of cellulose microfibrils in the secondary cell wall of tracheid
(Wimmer et al. 2002b). These microfibril’s function to provide wood elasticity and wood strength;
higher wood quality is usually associated with lower microfibril angles (angles more parallel to
the tracheid axis) (Huang et al. 2003). In conifers, MFA changes from bark to pith with smaller
angles found towards the bark (in mature wood) and larger angles toward the pith (juvenile wood)
(Van Leeuwen et al. 2011). The smaller the MFA, the higher the elasticity and the strength of the
wood (Donaldson 2008). MFA is controlled both genetically (Donaldson 2008) and
environmentally (Drew et al. 2013) and it seems that among environmental factors, temperature
and water availability have the strongest effect on MFA. Drew et al. (2013) also showed that
smaller MFA was attributed to summer temperature while larger MFA was correlated with spring
temperature, and MFA increased with rainfall and decreased with drought and water stresses.
2.2.3 Relationships between RW, EW, LW, and climate
16

Radial growth is the diametric growth of trees represented by annual-rings, which is controlled by
both genetic and environmental factors. However, radial growth is generally thought to be more
controlled by climatic factors than by genetic parameters (Downes and Drew 2008; Lenz et al.
2010); climate parameters showed higher correlation with radial growth, EW, and LW than genetic
influences did in white spruce (Picea glauca) (Lenz et al. 2010). Ring growth is shown to be
affected by both current growing season and previous growing season climate characteristics
(Fritts 1976).
The boreal forests in northern BC are limited in growth by temperature and long, cool winters
(Wood and Smith 2015). Hence, small changes in temperature in these regions could have large
proportional impact on wood properties. For examples, subalpine forests, adapted to cold
temperatures and large snowpack, may experience diminishing proportion of snowpack,
accelerated spring snowmelt, and expanding growth period leading to significant changes in radial
development (Wood and Smith 2015). In British Columbia in particular, fewer studies have been
conducted on relationships between climate variations as limiting factors and radial growth.
According to a study by Bouriaud et al. (2005), climate may have different impacts on tree-ring
traits including RW, EW, LW and density depending on species and stand conditions. In interior
central BC Wood et al. (2016) reported that temperature in early summer showed higher
correlation with RW formation of spruce stands than late summer temperature. Spletchtna et al.
(2000) found that radial growth of subalpine fir at higher elevations in BC was more impacted by
temperature than precipitation. they found that in general, impact of the temperature on radial
growth is more than that of precipitation. They also pointed out that tree-ring characteristics of
subalpine fir at high elevations were mostly correlated with summer temperature of current
growing season and with July temperature of current years as well as fall temperature of the
17

previous year. Brook et al. (1998), in a study on tree-ring characteristics of Black spruce (Picea
mariana (Mill.)) and Jack pine (Pinus banksiana Lamb.) in central Canadian boreal forests, found
that cool and wet climate is favorable for black spruce while high summer temperatures and spring
precipitation were favorable for jack pine. From different studies done on climate impact and forest
growth responses it can be inferred that response of tree anatomical traits to climate variables
depends on the species and site conditions.
2.2.4 Intra-annual wood characteristics vs annual-ring width
Intra-annual wood anatomical traits are generally more sensitive to change in climate variations
than tree-ring width. The higher correlations with climate records are particularly clear in wood
density and fibre properties (e.g., fibre length and microfibril angle) (Drew et al. 2013; Liang et
al. 2013). In stands where development is reliably limited by few climatic parameters, variation in
tree-ring widths is connected to these climatic restricting variables (Fritts 1976). However, in many
applications, RW estimations just give a growing season or annual portrayal of climatic
information not specific and/or short duration intervals like the beginning or end of growing
seasons (D’Arrigo et al. 1993). Furthermore, trees regularly do not grow at their upper edge limit
for examining climate impacts through RW. To measure and understand exact ecological changes,
even in small periods of time under changes in climate, intra-annually measured wood properties
including cell size (cell divider thickness, cell number), FL and FWT demonstrate higher
correlations to climatic variables than do RW orders (Wimmer and Grabner 2000; Davi et al.
2002). Considering that wood density is a significant component for anticipating end-use qualities
of wood, it is important to understand how wood density of timber species shifts with
environmental elements. For instance, in comparison to RW, MXD distinctively shows more
sensitivity (Davi et al. 2002) and is known to give altogether better intermediary records of
18

growing season conditions (D’Arrigo et al. 1992). These significant wood density-climate
correlations are credited to an increasingly conspicuous similarity between wood density and
climate from year to year.
2.2.5 Wood quality and need for switching to alternative species in BC
One of the essential requirements for sustainable timber management in Canada's future is
comprehension of the varieties and options in forests production across various locales and zones.
One zone could be the Engelmann Spruce-Subalpine Fir (ESSF) zone which has high timber,
watershed, wildlife, and recreational resource values (Hamann and Wang 2006). In interior British
Columbia, the ESSF zone represents a large and important timber supply (Hamann and Wang
2006). In this zone, subalpine fir (Abies lasiocarpa (Hook.) Nutt.)) along with Engelmann spruce
(Picea glauca x Engelmannii) are prevalent and dominant tree species in the continental, Rocky
Mountains of British Columbia, Canada (Meidinger and Pojar 1991). However, recently expanded
timber harvesting of spruce wood logging alongside other species like pine (Pinus sp.) as well as
expanded wildfires have diminished mature spruce availability in some lower-elevation habitats
in the Northern Rockies (Keane et al, 2018). Furthermore, because of the convergence of invasive
pests including mountain pine beetle (Dendroctonus ponderosae), spruce beetle (Dendroctonus
rufipennis) (Abatzoglou and Williams 2016), the wood supplies needed by the wood product
manufacturers are declining in interior BC. Therefore, forest managers need to think about some
alternative tree species to replace the above-mentioned species, and to guarantee sustainable forest
resources for future timber supply and silvicultural plans. It seems subalpine fir can be a proper
option. At higher elevation in the ESSF zone and more wetter areas, subalpine fir becomes the
dominant species (Alldritt-McDowell 1998). Along montane forest, central and subalpine forests of
British Columbia, subalpine fir, for example, is a major merchantable species (Splechtna et al.
19

2000). Of about four native fir species grown in Canada, subalpine fir has the widest growth range
in British Columbia (Jozsa 1991). Subalpine fir’s wood is odorless, light-weight, soft, and low in
bending and compressive strength which is easy to work, glues well, and holds nails and screws
fairly well (Alexander 1984). Its wood is primarily used for products such as lumber for home
construction and for prefabricated wood products having excellent pulping properties (Alexander
1984). Its long fibres and light color permit subalpine fir to be pulped well (Anderson 1956;
Alexander 1987). Overall, subalpine fir seems to be a good option to meet the wood demand of
British Columbia in the future, considering its adaptation to cold climates specially at higher
elevations.
2.2.6 Research Gap
This study aims at filling the gaps in knowledge on variation in subalpine fir wood anatomical
traits (EW, LW, RD, FL, FWT, MFA) under varying climate. Specifically, this study identifies the
regional interannual growth and development of subalpine fir at five latitudinally different study
sites and the relationship between growth and climate in these environments over time. Although
some dendroclimatic studies have been done to study subalpine fir wood properties in central and
southern BC (Klinka et al. 1992; Chen and Klinka 2000; Splechtna et al. 2000), still very few
studies have been completed studying subalpine fir’s wood anatomy responses to climate
parameters in northern BC (Zhang et al. 1999).
2.2.7 Objectives
This study aims to investigate how climate variables affect wood anatomy and growth of subalpine
fir at five stands along northern slopes of the Rocky Mountains in northern interior BC. The
objectives of this study are to: i) determine whether there are any statistically significant
20

relationships between climate variation and RW, EW, LW, MXD, MND, RD, FL MFA, FWT, and
if so, compare variation in subalpine fir’s wood properties to temperature and precipitation at five
geographically different sites in northern British Columbia, ii) Evaluate whether intra-annual wood
characteristics show stronger relationships with climate than do annual-ring widths, through
correlation analysis.
Hypothesis
I expect that subalpine fir trees would be responsive to climate in targeted locations and ring
density to be more sensitive to climate than ring width. Null hypothesis would be assuming there
is no statistically significant relationship between wood parameters and climate. So, if P value is
less than 0.05 then we reject the null hypothesis.
2.3 Method:
2.3.1 Site selection:
Subalpine fir trees were selected from five natural (not planted or previously harvested) sites
located along northern slopes, between 60 - 150 years old, at between 1120 to 1330 m above sea
level, in northern BC, Canada. Study sites selected, from northwestern most to southeastern most,
separated by latitude (55° 24ʹ to 53° 19ʹ N; 122° 32ʹ to 120° 07ʹ W) were: Pine Le Moray provincial
park, Mount Chingee, Mount Averil, Grizzly Den and McBride peak, respectively (Figure 2.1;
Table 2.1). The study sites were within a range of 450 to 480 km around Prince George city, BC,
Canada, from highest latitude site (Pine Le Moray) to the lowest one (McBride Peak) (Figure 2.1).
The reasons for choosing northern slopes as well as selecting higher altitudes were because
abundance of subalpine fir is higher on these site types (Keane et al. 1994), and it has been
suggested that subalpine fir is more vulnerable to climate change on northern slopes of the Rocky
21

Mountains than the southern ones (Keane et al. 2018). Additionally, at altitude above 1100 m,
subalpine trees become the dominant species (as observed) in comparison to spruce within the
Engelmann Spruce Subalpine fir zone (ESSF) and at higher elevations, trees are more sensitive to
climate variations, which are preferable for dendroclimatic studies.

Figure 2.1. Location of sampling sites for dendroclimatic study of subalpine fir along north-facing slopes of Rocky
Mountains in northern British Columbia. From top (highest latitude) to the bottom (lowest latitude): subalpine fir
sampled in Pine Le Moray provincial park, Mount Chingee, Mount Averil, Grizzly Den and McBride Peak.

22

Table 2.1: Chronology characteristics and geographic information for subalpine fir (Abies lasiocarpa) at five sampling sites for dendroclimatic study along northfacing slopes of the Rocky Mountains in northern British Columbia (DeLong et al. 1993; Wang et al. 2016).

Geographic information
Site Name

Pine Le
Moray

Chingee

Averil

Grizzly
Den

McBride
Peak

Elevation Longtitude
(m)
Latitude

1120

1220

1270

1250

1330

-122° 32ʹ
55° 24ʹ

-122° 53ʹ
55° 01ʹ

-122° 25ʹ
54° 24ʹ

-121° 30ʹ
53° 46ʹ

-120° 07ʹ
53° 19ʹ

Slope

N

N

NW

N

NW

Tree characteristics

Accumulative
Biogeoclimati Mean annual
Maximum Mimimum
annual
c zone/
temperature
teperature teperature Soil texture
precipitation
Subzone
(ºC)
Month
Month
(mm)

ESSF
Wet cold
(wc3)

ESSF
Wet cool
(wk2)

ESSF
Wet cool
(wk2)

ESSF
Wet cool
(wk1)

ESSF
Moist mild
(mm1)

1.5

0

1.3

2

1.6

925

1060

1045

1085

751

19.3
July

16.7
July

18
July

19.3
July

19.6
July

‑15.4 Medium to
January
coarse

‑15.3 Medium to
January
coarse

‑13.2
December

‑13.3
January

Fine to
coarse

Fine to
coarse

Indicator plants

Soil moisture Nutrient
regime
regime

goose berry 1
cranberry 2
devil's club 3
oak fern 4
mountain ash 5
thimble berry 6

Sub mesic
to
mesic
3-4

lady fern 7
gueens cup 8
black gooseberry 9
rhododendron 10
indian hellebore 11
huckleberry 12
gueens cup 8
oak fern 4
indian hellebre 11
five-leaf bramble 13
fireweed 14
black gooseberry 9
elderberry 15
black twinberry 16
rhododendron 10
lady fern 7
gooseberry 1
oak fern 4
fireweed 14

‑12.7
Mid coarse
17
one-side wintergreen
December
18

alder
thimbl berry 6

Core
Mean tree
length
Mean
age
(bark to
DBH (cm)
(years)
pith) (cm)

Poor to
medium
B-C

21.5

90

43

Mesi to sub Rich to
hydric
very rich
4-5
D-E

22.5

76

43

Poor to
Sub hydric
medium
5
B-C

18

104

36

Mesi to sub Poor to
hydric
medium
4-5
B-C

21

107

42

Poor to
medium
B-C

20.5

71

41

Mesic
4

1 Ribes uva-crispa; 2 Vaccinium subg. Oxycoccus; 3 Oplopanax horridus; 4 Gymnocarpium dryopteris; 5 Sorbus subg. Sorbus; 6 Rubus parviflorus; 7 Athyrium
filix-femina; 8 Clintonia uniflora; 9 Ribes lacustre; 10 Rhododendr3on albiflorum; 11 Veratrum viride; 12 Vaccinium membranaceum; 13 Rubus pedatus; 14
Chamaenerion species; 15 Sambucus racemose; 16 Lonicera involucrate; 17 Orthilia secunda; 18 Alnus sp.

23

All sample sites were within the ESSF zone, found at biogeoclimatic zone above 1000 m a.sl
(DeLong et al. 1993) ensuring that subalpine fir was the dominant species in study stands for
dendroclimatic studies to minimize interspecies competition. Cores at sampling sites were all taken
from the similar slopes (north-facing slope) to ensure sampling conditions were equal at different
sites. Biogeoclimatic variant of each site was determined with review of site characteristics and
Biogeoclimatic Ecosystem Classification land management handbooks (DeLong et al. 1993)
(Table 2.1).
2.3.2 Sample collection and preparation
At each stand, 22 dominant mature trees were sampled for statistical robustness (Anderegg et al.
2015). Subalpine fir trees were selected to the ecosystem relatively uniformly across soil,
vegetation and stand characteristic gradients, although every effort was made to minimize
heterogeneity across these characteristics. All selected trees experienced similar topographic
conditions (almost similar elevation and at similar slope (conditions) to ensure sample trees
possess similar tree-ring patterns in each stand (Fritts 1976). High elevated forests containing tall
and mature trees were selected to ensure climate was the most important factor affecting growth
(Fritts 1976). Basic criteria used in selecting trees at each site included choosing healthy subalpine
fir trees with open canopy (at least 5 meter in distance between two trees on the ground), and with
no fire or insect damage. The minimum inter-tree spacing was used to ensure growth was not
affected by competition, but mostly by climate variations (Fritts 1976). Since tree-ring
characteristics typically vary also vertically along the stem (Fritts 1976), breast height (1.3 m) was
used as a convenient reference height to control for this within-tree variation. Two 12 mm in
diameter core samples from opposite sides of trees were extracted at breast height from each tree
using increment borers. To minimize the effects of slope on tree-ring patterns, cores were extracted
24

from the cross-slope position, avoiding compression wood. GPS locations and site related data
were recorded at time of sampling. DBH was measured using core samples (by adding two samples
from each tree together if both met the pith otherwise, by doubling the length of the core samples
that reached the pith) at UNBC plant lab.
2.3.3 Fibre measurement
After extracting samples, tree cores were rolled in paper for transport. Once in the lab, cores were
air dried. One 12 mm in diameter core of each tree was used for analyzing fibre properties
including FL, FWT, MFA and FW. These cores were submerged in 95% Ethanol for 48 hours to
let available water in the cores be replaced with alcohol and therefore dry faster. Since tree resins
influence wood density (Lenz et al. 1976), dried core samples were placed in a Soxhlet chemical
extractor acetone system which cycled acetone over each sample for 5 hours (Jensen 2007) to
extract chemical compounds.
Increment cores were sanded on the radial side to make rings more visible, and core were
crossdated using a dotting system. Cores were labelled with a single dot every ten years, two dots
every half-century and three dots every century (Stokes and Smiley 1968); fine 600 grit sandpaper
was used to identify especially narrow rings. The Yamaguchi (1991) method was implemented to
determine significant marker years. Once cross-dated, twenty-two 12 mm diameter cores from
each site were cut into 10-year increment sections using scalpel blade. Then 10-year sections of
all trees were grouped together by year and corresponding years were analysed as an aggregate
sample (Table 2.2). Bark was cut off the cores and each group of similar 10-year sections were
weighed to target an optimum of 6.5 grams of biomass per 10-year interval (minimum of 5
samples) yielding composite samples A-K (Table 2.2). It worth mentioning that trees with almost
similar DBH and age class were targeted at sampling sites in order to minimize the effect of
25

differences in growth and differences in fiber size in younger trees comparing with mature trees.
Cores were then further analyzed at Canfor pulp’s head office, Burnaby, BC. Sections representing
each interval were put in 80 ml beakers and acetic acid and hydrogen peroxide (60%-40%) blend
were added to the beakers to completely submerge in the solution. This destructive solvent
prepares samples for PulpEye analysis by chemically breaking down the core sections, called a
delignification process, and turning them to cellulose bulk materials. PulpEye is a world-leading
modular analysis system developed for the pulp and paper industry. Its base unit consists of a
cabinet with space for three analyzer modules measuring pulp quality parameters online from one
or more sampling positions. Modules are combined and connected in a system providing analysis
data and control of any specific fibre property (Rahman 2018; Sundman et al. 2016).
Beakers were placed on oil bath at 70 °C leaving them for 48 hours (after the first 24 hours, core
sections of each beaker were mashed and squeezed). After the second day, samples were rinsed
thrice with distilled water to remove maceration solution and let air dry for 1 day in the fume hood.
Oven dry (OD) weights were then determined to get a net weight of 2.5 grams of each delignified
pulp sample. Samples were then rinsed in distilled water for 1 minute and poured into the PulpEye
system and results for FL, FW, FWT and MFA were reported as outputs in an Excel format. Fibre
property data including FL, FW, FWT and MFA were developed into chronologies by averaging
the series and differencing the mean dataset of each property to remove non-stationarity (e.g., 10year intervals and short time series) and were compared to climate variations (10-year averages),
using data from meteorological records obtained through ClimateBC, and to variation in the
Pacific Decadal Oscillation.

26

Table 2.2 Cutting, preparing, and labeling of 22 cores as aggregate, samples for further fibre analysis in Canfor Pulp
Innovation Center.

Section

Year
(Decade)

A
B
C
D
E
F
G
H
I
J
K

2019-2010
2009-2000
1999-1990
1989-1980
1979-1970
1969-1960
1959-1950
1949-1940
1939-1930
1929-1920
1919-1910

Chingee’s
core sections
weight (gr)

Pine’s core
sections
weight (gr)

Averil’s core
sections
weight (gr)

Grizzly’s
core sections
weight (gr)

McBride’s
core sections
weight (gr)

6.31
6.76
6.51
6.12
6.63
6.1
7.25
5.75

8.8
5.86
6.39
5.32
5.70
6.06
6.2
6.8
6.7
4.7
---

6.7
8.4
9.2
9.1
5.3
6.32
6.25
6.8
6.27
5.8
---

6.22
8.25
6.7
6.88
6.64
6.27
6.95
6.19
5.8
8.91
7.46

7.09
6.34
7.31
6.25
6.05
6.45
7.7
9.1
-------

-------

2.3.4 Preparation of cores for dendrochronological and densitometric analysis
The second 12 mm in diameter core taken per tree was used for densitometry and
dendrochronological analysis. Each core sample was mounted on a grooved mounting board
(Stokes and Smiley 1968) using wood glue and left to dry completely. After 2 to 3 days, mounted
cores were cut using a twin blade saw to obtain 2 mm laths revealing the radial surface of the core
(Haygreen and Bowyer 1996) for densitometry. Since wood chemical compounds like resin affect
density of wood (Grabner et al. 2005), an acetone Soxhlet apparatus was once again used to remove
wood resins from 2 mm lath samples (Grabner et al. 2005) prior to densitometry measurement.
Then, laths were measured using an ITRAX scanning densitometer in the Plant Lab at University
of Northern BC. ITRAX densitometry scan the whole lath length from bark to pith using an X-ray
laser beam. Measurements were made at 0.05 mm increments for 20 µs along each core and digital
x-ray images (radiographs) were analyzed using ITRAX Windendro® version (2009) to provide
27

maximum density (MXD) and mean density (MD) measurements.
2.3.5 Creating master chronology
I measured RW, EWW and LWW, and their boundaries using WinDendro image. Following ring
measurement, series were crossdated using WinDendro software by identifying characteristics
tree-ring patterns in graphical form. Visually cross-dated series were verified using COFECHA
(Holmes 1983) to ensure common end-dates, and the occurrence of false and missing rings
between trees were correct (Stokes and Smiley 1968). COFECHA matches individual time series
to a master chronology created from cores at each site (Grissino-Mayer 2001). By comparing each
core with the mean chronology of each site, the program generates the mean correlation between
the series (MSI). COFECHA also takes note of individual series that may contain errors such as
missed or false rings. The measurements are then validated until the remaining indicators
disappear or until the MSI is well above 0.42, the minimum value of significance at the 99%
confidence level (Grissino-Mayer, 2001). Cores with unique variation in ring width, not
representative of the stand, were eliminated (maximum of two to three cores) from the
chronologies for each site, allowing for common stand-level variation to be captured. Step
standardizations including detrending and stabilizing chronologies were then conducted to extract
the high frequency signal for tree RW measurements (Rossi 2003).
This process removes the non-climatic trends, such as the age-size related trends, and the effects
of stand dynamics (Fritts 1976) and removes growth-frequency variation and normalizing rapid
growth of young trees. I used the program ARSTAN to determine these variables, which applies
statistical analysis to develop a stand-level chronologies. The detrending conducted removed
growth-frequency variation attributed to tree geometry and normalized rapid growth of young trees
(Cook and Peters 1981; Meko and Baisan 2001) through the application of a negative exponential
28

curve to each series (Fritts 1976). In case it was not the best fit, then I applied Hugershoff curve to
remove biological trends. Raw values of annual RW, MD, EW and LW were then transformed
into dimensionless (index) values (Cook and Holmes 1999) for further analysis.
2.3.6. Comparing chronologies with climate
Climate data
Mean monthly and seasonally averaged air temperature, and mean accumulated precipitation were
obtained from ClimateBC at https://cfcg.forestry.ubc.ca/projects/climate-data/climatebcwna/.
ClimateBC calculates seasonal and annual climate variables for specific locations based on
latitude, longitude, and elevation. The program uses a combination of bilinear interpolation and
elevation adjustments to predict climate data from 1900 to 2018. Seasonal averages in ClimateBC
include winter (previous December, current January, and February), spring (March, April and
May), summer (June, July and August), and autumn (September, October and November). The
Pacific Decadal Oscillation (PDO) is often described as a long-lived El Niño-like pattern of Pacific
climate variability (Zhang et al. 1997). The severe phases of the PDO have been classified as being
either warm or cool. When sea level pressures are below average over the North Pacific, the PDO
has a positive value and conversely, above average sea level pressures over the North Pacific,
result in a negative value in PDO (Mantua 1999). PDO could affect climate regimes in western
Canada during its warm and cool phases. Pacific Decadal Oscillation records were obtained from
National Centers for Environmental Information or National Oceanic and Atmospheric
Administration (NOAA) at https://www.ncdc.noaa.gov/teleconnections/pdo/
2.3.7. Correlation and regression analysis between climate and growth parameters
RW, EW, LW, MD and RD of all 5 sites were tested for normality using IBM SPSS version 26.
29

Pearson’s correlation coefficients were calculated for standardized, detrended master chronologies
and climate variations to determine significant correlations. Spearman’s correlation coefficient
was calculated for non-normal dependant variables in SPSS.
Standardized detrended fibre properties including FL, FW, FWT and MFA were compared with
mean monthly, annual and seasonal air temperature, mean monthly maximum and minimum air
temperature, and total annual and seasonal precipitation using a Pearson’s correlation coefficient.
Normality of FL, FWT, MFA and FW were tested using Shapiro-Wilk test in SPSS (because the
number of data and sample size were small) before correlation analysis.
2.3.8 Regression Analysis and Comparison of Measured and Modelled Values
Regression analysis was completed where strong, significant Pearson’s or Spearman’s correlation
coefficients were identified, above 0.4 (Wood and Smith 2012; Blanchette et al. 2015) (Tables 2.5,
2.6, 2.9). Regression models with significant correlation coefficients, and strong R2 values (>0.25)
were visually assessed against measured values to determine model accuracy over time. To obtain
stronger models, multi variable model can be applied using two or more climatic variables and
putting in the same model to predict better and stronger models provided that the independent
variables (climatic parameters) are not statistically correlated. To find if variables were related
together or not, a Durbin-Watson statistic was applied. Durbin Watson values are between 0 and
4. A value of 2.0 means that there is no autocorrelation detected in the sample. Values from 0 to
less than 2 indicate positive autocorrelation and values from 2 to 4 indicate negative
autocorrelation. If independent variables were autocorrelated we could potentially take an average
of two independent variables and put that average in the model for prediction. For those
independent climate variables that were basically different (e.g., temperature and precipitation),
and needed to be combined in a model, fluctuations over time were compared visually to make
30

sure they have same trends (See Figure A1, A2, A3, A4 in Appendix). Split verification was then
used to validate the models. The latter 50% of time series was applied to calibrate the models. All
predicated site-specific anatomical traits passed the verification test at 90% significance of
reduction of error values (RE) compared to actual measured data.
2.4 RESULTS
2.4.1 Changes in Climate
Meteorological records for five study sites around Prince George are shown in Figures 2.2 and 2.3.
Figures 2.2 - 2.3 depict mean annual temperature and total annual precipitation from 1940 to 2018.
Annual precipitation has generally decreased at the Pine Le Moray and Mt. Chingee sites over
time and increased at McBride Peak since 1920 (Figure 2.3 A, B, E). At the Grizzly Den and Mt.
Averil sites, totals of annual precipitation generally show steady trends since 1940 (Figure 2.3 C,
D).

31

Mean Air Temperature ºC
Figure 2.2 Mean annual temperature for five study sites from 1940-2018. Average temperature has generally
increased at study sites over time since 1940. A= Pine Le Moray; B=Mount Chingee; C=Mount Averil; D=Grizzly
Den; E=McBride Peak. The dashed average represents a 10-year moving average in temperature variation. Thin
smooth lines represent trendlines during 1940- 2018 period.

32

Precipitation (mm)
Figure 2.3 Total annual precipitation for five study sites from 1940-2018. Annual precipitation has generally decreased
at the Pine Le Moray and Chingee sites (A, B) over time while increased at McBride since 1920 (E). At the Grizzly
Den and Averil sites, total annual precipitation generally shows steady trends since 1940 (C, D). A= Pine Le Moray;
B = Mount Chingee; C = Mount Averil; D = Grizzly Den; E = McBride Peak. A 10-year moving average line in
precipitation variation is shown with dashed line. Thin smooth lines represent trendlines during 1940 – 2018 period.
Note not all y-axis are the same.

33

2.4.2 Correlation between climate and Fibre properties
Several significant correlations (p < 0.05) were found between tracheid (fibre) characteristics and
the previous and current monthly and seasonal temperature and precipitation records of Pine Le
Moray, Mt. Chingee, Mt. Averil, Grizzly Den and McBride Peak sites (Figure 2.4 – 2.7; Table 2.5
– 2.6). FL was positively correlated with climate indices at Grizzly Den and McBride Peak sites
while negatively correlated with Averil site. FWT showed positive correlations with climate
variables at Pine Le Moray, Mt. Change and McBride Peak sites. MFA and FW were both
positively correlated with climate parameters at Pine Le Moray and McBride Peak sites while both
properties showed negative correlations with climate at Mt. Chingee site (Table 2.5). All fibre data
(FL, FW, FWT and MFA) were deemed normal (Table 2.4). Durbin-Watson statistic tests showed
that independent climate variables were related to each other (See Table A1 in Appendix).
Therefore, we took average of corresponding variables for comparing with fibre properties and for
modeling.

34

Table 2.3. Normality statistics for ring width (RW), earlywood density (EWD), earlywood width (EWW), latewood density LWD, latewood width (LWW),
maximum density (MXD), minimum density (MND) and ring density (RD) in Pine Le Moray, Chingee, Averil, Grizzly Den and McBride sites including Skewness
and Kurtosis values. Skewness and Kurtosis > +1.96 or < -1.96 were not accepted as normal data. Bold values indicate non- normal values.

Site

RW

EWD

EWW

LWD

LWW

MXD

MND

RD

Skewness Kurtosis Skewness Kurtosis Skewness

Kurtosis Skewness Kurtosis Skewness Kurtosis Skewness Kurtosis Skewness Kurtosis Skewness Kurtosis

Pine Le Moray

0.068

-0.777

1.299

0.834

0.393

-0.811

-0.061

-1.457

1.381

0.345

0.171

Mt Chingee

0.023

-1.909

0.061

-0.787

-0.050

-1.908

0.593

-0.551

0.843

0.739

Mt Averil

3.110

1.196

1.995

0.416

2.823

1.103

0.040

-0.949

Grizzly Den

-0.563

-0.835

1.214

0.152

0.030

-0.177

1.214

McBride Peak

-0.410

-0.974 -0.532

-0.795

-0.696

-1.260

1.230

-0.248 0.896

-1.494 -0.306

0.002

-0.441 -1.588 1.231

-0.346

0.477

-0.438

-1.886 -0.517

0.217

-1.100 3.523

2.861

0.003

-0.609

0.152

-2.376

0.402

-2.599 -0.227 1.416

0.014

0.767

-1.164

-1.123

0.338

-0.708 -0.717 -0.841 1.540

-0.900 -0.746 -0.892

35

Table 2.4. Normality statistics for fibre length (FL), fibre wall thickness (FWT), micro fibril angle (MFA) and fibre
width (FW) in Pine Le Moray, Mt. Chingee, Mt. Averil, Grizzly Den and McBride Peak sites including Shapiro-Wilk
test to validate normality of sites with small fibre sample sizes. P-value> +0.05 is considered as normal data.

FL

Site

FWT

MFA

FW

Shapiro-Wilk Sig (P-value ) Shapiro-Wilk Sig (P-value ) Shapiro-Wilk Sig (P-value ) Shapiro-Wilk Sig (P-value )

Pine Le Moray

0.888

0.23

0.943

0.64

0.915

0.39

0.935

0.56

Mt. Chingee

0.957

0.79

0.861

0.15

0.906

0.37

0.964

0.85

Mt. Averil

0.913

0.34

0.960

0.80

0.920

0.39

0.877

0.15

Grizzly Den

0.919

0.35

0.926

0.41

0.958

0.76

0.960

0.78

McBride Peak

0.821

0.07

0.925

0.51

0.866

0.17

0.848

0.12

Table 2.5. Pearson correlations between climate variations and fibre length (FL) and fibre wall thickness (FWT) for
Pine Le Moray, Mt. Chingee, Mt. Averil, Grizzly Den and McBride Peak sites. Only strongest and significant
correlations have been shown. Only R > 0.7 was accepted as significant correlation. “⁎” Values are significant at 95%
(P<0.05) and “⁎⁎” shows significance at 99% (P<0.01). Oct = October.

Climatic parameter

Fiber Length
Chingee Averil

Pine

Fiber Wall Thickness

McBride Grizzly Chingee Averil Pine

McBride Grizzly

-0.512 -0.15 0.161 0.454 0.611 -0.11 0.757* -0.131
Maximum Temperature (July)
0.682 -0.608 -0.32 0.571 0.044 0.802* -0.46 -0.131 -0.475
Minimum Temperature (Previous_Oct) -0.182 0.482 0.195 -0.185 0.029 -0.329 -0.29 0.273 0.867*
Maximum Temperature (Fall + summer) 0.506 -0.763* -0.028 -0.089 0.131 0.371 -0.42 -0.018 -0.692

Mean + Maximum Temperature (May)

0.511

Mean and Minimum temperature of
December

0.339

0.383

0.412

Minimum Temperature and
Precipitation (Previous Fall)

-0.411

0.234

0.68 0.883** 0.043 0.381 -0.41 0.218

-0.29 0.713*

0.3

-0.31 0.139 -0.538
0.126

0.121
-0.274
0.441
-0.202
0.579
0.041

36

Table 2.6. Correlations between climate variations and, micro fibril angle (MFA) and fibre width (FW) for Pine Le
Moray, Mt. Chingee, Mt. Averil, Grizzly and McBride Peak sites. Only strongest and significant correlations (R >
0.7) are accepted as significant correlation. “⁎” Values are significant at 95% (P<0.05) and “⁎⁎” show significance at
99% (P<0.01).
Climatic parameter

Micro Fibril Angle
Pine Le
Chingee
Moray

Fibre Width

McBride Pine Le
Averil Grizzly
Chingee
Peak
Moray

Averil Grizzly

McBride
Peak

Precipitation (Previuos October)

-0.491 -0.199 0.109 0.284 0.331
0.2
0.525 -0.305 0.145 -0.972**
-0.216 -0.807* 0.046 0.142 0.038 0.266
0.58
-0.18 -0.14 -0.577
0.691 -0.291 -0.137 -0.175 0.819* 0.418 -0.044 0.23
0.154 0.247
-0.171 -0.237 0.282 -0.111 0.572 -0.251 -0.812* 0.024 -0.093 0.028

Mean Temperature (August) and
Minimum Temperature (July + August)

0.867** -0.217 -0.231 -0.066

0.496

0.679

0.111

0.18

0.337

-0.042

0.697

0.228

0.85** 0.309

0.28

0.35

0.211

Maximum Temperature (fall)
Maximum Temperature (July)
Minimum Temperature (August)

Mean Temperature (June + July) and
Minimum Temperature (July)

-0.483 -0.202

-0.38

FL at the Mt. Averil site was significantly and negatively correlated (R = -0.763) with maximum
air temperature of Fall and summer (Figure 2.4; Table 2.5), while this property at McBride Peak
and Grizzly Den sites showed significant positive correlations (R = 0.883 and 0.713) with
minimum temperature combined with precipitation in previous year fall season, and average
temperature combined with minimum temperature of December respectively (Figure 2.4; Table
2.5).
Results showed maximum July temperature was significantly and positively (R = 0.802) related
with FWT at Mt. Chingee during 1950- 2018. Maximum and average May temperature was
positively correlated (R = 0.757) with FWT at Pine Le Moray during 1940- 2018. FWT was
correlated (R = 0.867) with minimum temperature of the previous October at McBride Peak from
1950 to 2018 (Figure 2.5; Table 2.5).

37

Figure 2.4 Relationships between climate variables and fibre length (FL) for Grizzly Den, McBride Peak and Averil
sites during 1920 to 2018. A: Relationship between average and minimum temperature of December and FL (Left),
and linear relation of average and minimum temperature of December and FL at Grizzly site during 1920-2018 (Right).
B: Relationship between minimum temperature and precipitation of previous fall and FL (Left), and linear relation of
minimum temperature and precipitation of previous fall and FL at McBride site during 1950-2018 (Right). C:
Relationship between maximum temperature of fall and summer and FL (Left), and linear relation of maximum
temperature of fall and summer and FL at Averil site during 1930- 2018 (Right). Tmean= mean temperature, Tmax=
maximum temperature, Tmin= minimum temperature, Ave= average, Pre= previous, PPT= precipitation, STNDRD=
standardized, sm= summer, FL= fibre length.

38

Figure 2.5 Relationships between climate variables and fibre wall thickness (FWT) at Pine Le Moray, Chingee and
McBride peak sites during 1940 to 2018. A: Relationship between average and maximum temperature of May and
FWT (Left), and linear relation of Average and maximum temperature of May and FWT at Pine Le Moray during
1940-2018 (Right). B: Relationship between maximum temperature of July and FWT (Left), and linear relation of
maximum temperature of July and FWT at Chingee during 1950- 2018 (Right). C: Relationship between minimum
temperature of previous October and FWT (Left), and linear relation of minimum temperature of previous October
and FWT at McBride during 1950-2018 (Right). Tmean= mean temperature, Tmax= maximum temperature, Tmin=
minimum temperature, Ave= average, Pre= previous, PPT= precipitation, STNDRD= standardized, sm= summer,
Oct= October.

39

Some significant correlations were obtained from comparison between climate variables and MFA
at study sites. Figure 2.6 shows the relationships between climate variables and MFA for Pine Le
Moray, Mt. Chingee and McBride Peak from 1940 to 2018. Maximum July temperature of July
was significantly and negatively (R = -0.807) correlated with MFA at Chingee. At Pine Le Moray,
a combination of average temperature of August and minimum temperature of July and August
were positively correlated (R = 0.867) with MFA. August minimum temperature was positively
and significantly correlated with MFA at McBride site (Figure 2.6; Table 2.6).
The relationships between climate variables and fibre width at Pine Le Moray, Mt. Chingee and
McBride Peak from 1940 to 2018 are shown in Figure 2.7. Previous October precipitation was
negatively (R = -0.812) correlated with FW at Mt. Chingee site. Results for Pine Le Moray site
showed positive correlation (R = 0.85) between average June and July temperature in combination
with minimum July temperature, and FW.

40

Figure 2.6 Relationship between temperature and micro fibril angle (MFA) at Pine Le Moray, Chingee and McBride
Peak sites during 1940 to 2018. A: Relationship between Average temperature of August combined with minimum
temperature of July and August, and MFA (Left), and linear relationship of Average temperature of August combined
with minimum temperature of July and August, and MFA at Pine Le Moray during 1940-2018 (Right). B: Relationship
between maximum temperature of July and MFA (Left), and linear relation of maximum temperature of July and
MFA at Chingee during 1950- 2018 (Right). C: Relationship between minimum temperature of August and MFA
(Left), and linear relation of minimum temperature of August and MFA at McBride Peak during 1950-2018 (Right).
Tmean / ave= mean temperature, Tmax= maximum temperature, Tmin= minimum temperature, STNDRD=
standardized, sm= summer, Jul = July, Aug = August, Sept = September.

41

Figure 2.7 Relationships between temperature and precipitation, and fibre width (FW) at Pine Le Moray, Chingee and
McBride Peak sites during 1940 to 2018. A: Relation between average temperature of June and July combined with
minimum temperature of July, and FW (Left), and linear relation of average temperature of June and July combined
with minimum temperature of July, and FW in Pine Le Moray during 1940-2018 (Right). B: Relationship between
previous October precipitation and FW (Left), and linear relation of previous October precipitation and FW at Chingee
during 1950- 2018 (Right). C: Relationship between fall maximum temperature and FW (Left), and linear relation of
fall maximum temperature and FW at McBride Peak during 1950-2018 (Right). Tmean / ave= mean temperature,
Tmax= maximum temperature, Tmin= minimum temperature, PPT = precipitation, pre = previous, STNDRD=
standardized, sm= summer, Oct = October, Jun = June, Jul = July.

42

Chronologies
Chronology lengths for study sites were between 69-103 years at Pine Le Moray, 55-95 at Mount
Chingee, 68-138 at Mount Averil, 55-148 at Grizzly Den and 56-86 at McBride Peak (Table 2.7).
The year of residual series cut-off was variable for radial growth characteristics and was based on
accepted minimum number of trees with common starting year at each site (Table 2.7). At each
site, 2-4 chronology series were disregarded from further analysis due to unsuccessful cross-dated
series (Table 2.7). Normality of RW, EW, LW, EWD, LWD, MD and ring density chronologies
were tested using SPSS before correlation analysis. Results determined most data was normal with
the exception of Averil-RW, Averil-EWW, Averil-EWD, Averil-MND, Grizzly-LWW and
Grizzly-MXD (Table 2.3).
2.4.3 Climate Correlations
Several significant correlations were found from comparisons of winter (October to March) and
spring (April to June) circulation indices - Pacific Decadal Oscillation (PDO) and mean winter and
spring temperatures at study sites (Table 2.8; Figure 2.21- 22). Winter PDO was significantly and
positively correlated with winter temperature at Pine Le Moray, Mt. Chingee, Mt. Averil, Grizzly
Den and McBride Peak sites at 99% confidence (Table 2.8). Spring PDO was significantly and
positively correlated with spring temperature at Averil and McBride sites (Table 2.8; Figure 2.22).
In general, PDO cooling phase during 1940s to 1970s matched with local climate trends at study
sites. Similarly, local climate variations at study sites during 1980s to 1990s followed PDO
warming phase during 1980s to 1990s (Figure 2.21- 2.22).

43

Table 2.7. Master chronology statistics for study stands (Pine Le Moray, Chingee, Averil, Grizzly Den and McBride
Peak) along north-facing slopes of Rockies in northern interior British Columbia. ring width (RW), earlywood width
(EWW), earlywood density (EWD), latewood width (LWW), latewood density (LWD), maximum density (MXD),
minimum density (MND) and ring density (RD). MSI = Mean series intercorreltion: indicates the strength of
correlation between samples taken from the same location, based on 50-year overlapping segments.
# of Trees
Cut off
Mean Length
Accepted
Chronology
Chronology
Successfully
Site Name
MSI
series length
of Series
series
Type
Length
Crossdated
(start year)
(year)
Length

Pine Le
Moray

Mt. Chingee

Mt. Averil

Grizzly
Den

McBride
Peak

RW
EWW
EWD
LWW
LWD
MXD
MND
RD
RW
EWW
EWD
LWW
LWD
MXD
MND
RD
RW
EWW
EWD
LWW
LWD
MXD
MND
RD
RW
EWW
EWD
LWW
LWD
MXD
MND
RD
RW
EWW
EWD
LWW
LWD
MXD
MND
RD

0.563
0.612
0.643
0.559
0.711
0.686
0.531
0.51
0.664
0.596
0.642
0.701
0.612
0.559
0.614
0.771
0.744
0.696
0.723
0.632
0.753
0.713
0.651
0.618
0.693
0.722
0.634
0.748
0.761
0.663
0.629
0.592
0.743
0.777
0.729
0.614
0.672
0.662
0.779
0.726

1915-2018

1940

85

80

19

1923-2018

1950

67

70

18

1880-2018

1940

101

80

19

1870-2018

1940

101

80

18

1932-2018

1960

64

60

18

44

Table 2.8. Relationships between circulation indices including winter (October – March) and spring (April-June) and
temperature of winter and spring in study stands (P, C, A, G, M). PDO- Pacific Decadal Oscillation. All sites are
Significant at 0.99 (P < 0.01).

Site

Longitude

Latitude

Elevation
(a.s.l)

Slope

Pine Le Moray
Mt. Chingee
Mt. Averil
Grizzly Den
McBride Peak

55° 24ʹ
55° 01ʹ
54° 24ʹ
53° 46ʹ
53° 19ʹ

-122° 32ʹ
-122° 53ʹ
-122° 25ʹ
-121° 30ʹ
-120° 07ʹ

1120
1200
1270
1250
1330

N
N
NW
N
W

PDO_winter PDO_spring
sig
2
2
(P value)
(R )
(R )

0.268
0.283
0.308
0.308
0.302

0.326
0.342

<0.0001
<0.0001
<0.0001
<0.0001
<0.0001

2.4.3.1 Relation between climate parameters (temperature and precipitation) and RW, EW,
LW, MXD, MND, RD
Some significant correlations (p < 0.05) were found between subalpine fir’s tree-ring
characteristics (RW, EW, LW, MXD, MND, RD) and the previous and current monthly and
seasonal temperature and precipitation of Pine Le Moray, Mt. Chingee, Mt. Averil, Grizzly Den
and McBride Peak sites. (Figures 2.8 - 2.13). Results show that intra-annual wood characteristics
including (EW, LW, RD) have stronger and more significant correlations with climate variables
than do annual-ring width. In general, because of the availability of statistically significant
correlations (P < 0.05) between subalpine fir radial growth and climate, the null hypothesis is
rejected.

45

Correlation Coefficient

Months of the year
Figure 2.8. Correlations between mean monthly temperature and precipitation at Pine Le Moray, Mt. Chingee, Mt. Averil, Grizzly Den and McBride Peak stands
and ring width (RW), earlywood width (EWW) along north-facing slopes of the Rockies in northern BC. Significant correlations are depicted by ** for p < 0.01
and * for p < 0.05. Months are represented by lower case (previous year) and upper case (current year) letters, respectively; WT = winter, SP= spring, SM= summer,
AT = fall. S = Spearman’s correlation coefficient.

46

Correlation Coefficient

Months of the year
Figure 2.9. Correlations between mean monthly temperature and precipitation at Pine Le Moray, Mt. Chingee, Mt. Averil, Grizzly Den and McBride Peak stands
and earlywood width (EWW), latewood width (LWW) along north-facing slopes of the Rockies in northern BC. Significant correlations are depicted by ** for p
< 0.01 and * for P < 0.05. Months are represented by lower case (previous year) and upper case (current year) letters, respectively; WT = winter, SP= spring, SM=
summer, AT = fall. S= Spearman correlation coefficient.

47

Correlation Coefficient

Months of the year
Figure 2.10. Correlations between mean monthly temperature and precipitation at Pine Le Moray, Chingee, Averil, Grizzly Den and McBride Peak stands and
earlywood density (EWD) and latewood density (LWD) along north-facing slopes of the Rockies in northern BC. Significant correlations are depicted by ** for p
< 0.01 and * for p < 0.05. Months are represented by lower case (previous year) and upper case (current year) letters, respectively; WT = winter, SP= spring, SM=
summer, AT = fall. S = Spearman correlation coefficient.

48

Correlation Coefficient

Months of the year
Figure 2.11. Correlations between mean monthly temperature and precipitation at Pine Le Moray, Chingee, Averil, Grizzly Den and McBride Peak stands and
latewood density (LWD) and maximum density (MXD) along north-facing slopes of the Rockies in northern BC. Significant correlations are depicted by ** for
P < 0.01 and * for P < 0.05. Months are represented by lower case (previous year) and upper case (current year) letters, respectively; WT = winter, SP= spring,
SM= summer, AT = fall. S = Spearman correlation coefficient.

49

Correlation Coefficient

Months of the year
Figure 2.12. Correlations between mean monthly temperature and precipitation at Pine Le Moray, Chingee, Averil, Grizzly Den and McBride Peak stands and
minimum density (MND) and ring density (RD) along north-facing slopes of the Rockies in northern BC. Significant correlations are depicted by ** for P < 0.01
and * = 0.05. Months are represented by lower case (previous year) and upper case (current year) letters, respectively; WT = winter, SP= spring, SM= summer,
AT = fall.

50

Correlation Coefficient

Months of the year
Figure 2.13. Correlations between sum of annual precipitation at Pine Le Moray, Chingee, Averil, Grizzly Den and McBride Peak stands and ring density along
north-facing slopes of the Rockies in northern BC. Significant correlations are depicted by ** for P < 0.01 and * for P < 0.05. Months are represented by lower
case (previous year) and upper case (current year) letters, respectively; WT = winter, SP= spring, SM= summer, AT = fall.

51

2.4.4 Regression Analysis and Comparison of Measured and Modelled Values
Results of Durbin Watson statistics showed that some independent variables that are used to make
the models stronger, were related. Therefore, we took an average of two or more climate variables
and put them in the model for stronger predictions at each site (Pine Le Moray, Averil and McBride
Peak sites) (See Figure A1, A2, A3, A4 in Appendix). Although there were numerous significant
relationships found between climate, and radial and fibre properties growth in the study sites (Pine
Le Moray, Chingee, Averil, Grizzly Den, McBride Peak), the models presented reflect the
strongest relationships found between wood properties and monthly and seasonal climate (Tables
2.9 and Figures 2.14 – 2.20).

2.4.4.1 Relationships between fibre properties (FL, FWT, MFA and FW) and climate
McBride Peak

All four measured fibre properties (FL, FWT, MFA and FW) at McBride Peak were successfully
modeled from climate variations. Similarity between modeled FL obtained from combination of
previous fall precipitation and previous fall minimum temperature and measured FL, was high
(robust) between 1950 and 2018, with some decoupling between the observations and predicted
models from 1980s to 1990s (Figure 2.14 A). The predicted model of FWT from the McBride
Peak site, based on minimum temperature of previous October, corresponded well with measured
data (Figure 2.14 B). Measured values of MFA from McBride Peak and modeled values of MFA
from September minimum temperature decoupled in 1960s, 1970s and 2000s (Figure 2.14 C).
Modelled values of the FW based on fall maximum temperature followed the measured values
strongly during 1950-2018 period (Figure 2.14 D).

52

Standardized FL
Standardized FWT
Standardized MFA
Standardized FW

Time (decade)
Figure 2.14 Measured fibre properties (solid line) vs modelled fibre properties (dotted line); dashed lines indicate 1
decade lagged data fibre length (FL), fibre wall thickness (FWT), micro fibril angel (MFA) and fibre width (FW) for
subalpine fir at McBride Peak stand in northern British Columbia. Data modeled from combined previous fall
precipitation with previous fall minimum temperature (A), minimum temperature of previous October (B), September
minimum temperature (C), and previous fall maximum temperature (D). R2 values are presented for ** p <0.01 and *
p < 0.05. Note not all axis are the same scale.

53

Pine Le Moray

FWT, MFA and FW were successfully modeled from climate values at the Pine Le Moray site
during 1940-2018. Modeled FWT values were obtained from average and maximum temperature
of May. These modelled values were decoupled from measured values in the 1970s, and 2000s.
However, the highest correlation between model and measured FWT values occurred between the
1940s and 1960s (Figure 2.15 A). Correlations were high between modeled MFA values obtained
from combined average temperate of August with minimum temperature of July and August and
measured MFA values for the decades between 1950s, 1960s and 1980s, and low in1970s and
1990s (Figure 2.15 B). The highest correlation between modeled RW values obtained from
combined average temperature of June and July with minimum temperature of July and measured
RW values occurred in 1980s. The lowest correlations for the RW model and measured values
occurred in the 2010s (Figure 2.15 C).

54

Standardized FWT
Standardized MFA
Standardized FW

Time (decade)
Figure 2.15 Measured fibre properties (solid line) vs modelled values (dotted line); dashed line indicates 1 decade
lagged data for fibre wall thickness (FWT), micro fibril angel (MFA) and fibre width (FW) for subalpine fir at Pine
Le Moray stand in northern British Columbia. Data modeled from average and maximum temperature of May (A),
combined average temperate of August with minimum temperature of July and August (B), and combined average
temperature of June and July with minimum temperature of July (C). R 2 values are presented with ** p <0.01 and * p
< 0.05 Note not all axis are the same scale.

55

Mt. Chingee
FWT, MFA and FW were successfully modeled from climate at the Chingee site. The predicted
model for FWT based on July maximum temperature followed measured values well, with highest
correlation between observed and predicted models in the 2000s and 2010s and lowest in the 1970s
(Figure 2.16 A). The highest correlations between measured values and the predicted model for
MFA calculated from July maximum temperature at Chingee site occurred in the 1970s, 1990s and
2000s while the lowest correlations occurred in the 1980s (Figure 2.16 B). During the 1960s to
1970s and 2000s, the predicted model for FW obtained from previous October precipitation
strongly followed measured values. However, the weakest relationship between modeled and
measured RW values occurred in the 1990s (Figure 2.16 C).
Mt. Averil
Although FL was successfully modeled from combined maximum temperature of summer and
previous fall at the Averil site during 1930 to 2018, the models had some areas of poor association
with measured values in 1960s, 1980s, 1990, and 2010s (Figure 2.17 A). Other fibre characteristics
(FW, FWT and MFA) did not show high correlations with the most important climate variables
(temperature and precipitation) models.
Grizzly Den
Like Averil site, none of FWT, MFA or FW values showed significant correlation with climate
variables from the Grizzly site. Only FL values were successfully modeled from combined
December mean and minimum temperatures during 1920 - 2018 (Figure 2.17 B). The strongest
similarities between observed values and predicted model values occurred in the 1920s and 1930s.
However, measured values and predicted model values decoupled between 1960s to 2010s (Figure
2.17 B).
56

Standardized FWT
Standardized MFA
Standardized FW

Time (decade)
Figure 2.16 Measured fibre properties (solid line) vs modelled values (dotted line); dashed lines indicate 1 decade
lagged data in fibre length (FL), fibre wall thickness (FWT), micro fibril angle (MFA) and fibre width (FW) for
subalpine fir at Chingee site in northern interior British Columbia. Data modeled from July maximum temperature
(A), July maximum temperature (B), and previous October precipitation (C). R2 values are presented with ** p <0.01
and * p < 0.05. Note not all axis are the same scale.

57

Standardized FL
Standardized FL

Time (decade)
Figure 2.17 Measured fibre properties (solid line) vs modelled values (dotted line); dashed lines indicate 1 decade
lagged data for fibre length (FL) at Averil and Grizzly Den stands in northern interior British Columbia. Data modeled
from combined maximum temperature of summer and fall (A), and combined December mean and minimum
temperatures (B). R2 values are presented with ** p <0.01 and * p < 0.05. Note not all axis are the same scale.

2.4.4.2 Relationship between temperature and RW, EWD, MND, EWW, MXD, RD
Like the fibre property correlations reported, only strongest correlations between RW, EWD,
MND, EWW, MXD, RD and climate (R2 > 0.2; r > 0.447) were considered for modeling radial
growth properties (Tables 2.9 and Figures 2.18 – 2.20).

58

Table 2.9. Regression analysis between climate variables (precipitation and temperature) and Earlywood width
(EWW), earlywood density (EWD), minimum density (MND), ring width (RW), latewood density (LWD) and ring
density (RD) chronologies for Pine, Chingee, Averil, Grizzly and McBride stands along north-facing slopes of the
Rockies in northern BC. R2 values only accepted if > 0.2. All values are significant at 0.99. Combined temperature
and precipitation = precipitation of May, June, July and temperature of September, October, November, and
December; Combined precipitation and temperature = precipitation of May, September, October, November and
December, and temperature of May and June.
EWW EWD
(R2)
(R2)

MND
(R2)

RW
(R2)

LWD
(R2)

RD
(R2)

Climatic
Parameter(s)

Site

August temperature & September precipitation

Pine Le Moray

0.214

<0.0001

Summer temperature

Mt. Chingee

0.331 0.321

<0.0001

Combined temperature & precipitation

Mt. Averil

Combined precipitation & temperature

Mt. Averil

April temperature

Grizzly Den

0.243

<0.0001

Ausut temperature

McBride Peak

0.272

<0.0001

Previous July & August temperature

McBride Peak 0.369

sig
(P value)

0.202 <0.0001
0.203

<0.0001

<0.0001

Mt. Chingee

EWD and MND chronologies for the Chingee site were successfully modelled from current
summer temperature and showed significant relationships to measured values over the time series
with the strongest correlations from 1957 to 1958, 1965 to 1969, 1989 to 1990 and 1995 to 1995
(Figure 2.18 A). However, EWD and MND models were poorly correlated with measured values
from 1973 to 1975, 1983 to 1987 and after 2002 (Figure 2.18 A).
McBride Peak
In general, MND, EWW and RW for subalpine fir in McBride site were strongly correlated with
August temperature, a combination of previous July and August temperature, and previous July
temperature, respectively (Figure 2.18 B, C; Figure 2.19 A). Correlation was high between
measured MND values and modeled values obtained in McBride site for the years between 1979
59

to 1982. However, measured MND value showed low correlation with modelled values between
2001 and 2002 (Figure 2.18 B). The second time period with the highest correlation between
measured and modeled MND was between 1967 to 1968 (Figure 2.18 B).
EWW values from the McBride site were modeled from combined previous July and August
temperature. This model showed low correlation with measured EWW from 1968 to 1980 (Figure
2.18 C). The modeled chronology for RW values from the McBride site, obtained from previous
July temperature, was generally more strongly correlated with measured RW after 1980 compared
to prior to 1980 (Figure 2.19 A). The strongest correlation between modeled and measured RW
was observed from 1987 to 1989. However, the weakest relationship between modeled and
measured RW was between 1975 to 1980 (Figure 2.19 A).
Grizzly Den
Subalpine fir MND chronology from the Grizzly Den site was modelled from current April
temperature. This model showed weak to moderate correlation with measured values (Figure 2.19
B). Some similarities between the modeled and measured chronologies included: 1945- 1946,
1968-1969, 1975-1976 and 2012-2013 (Figure 2.19 B).

60

Standardized EWD
Standardized MND
Standardized MND
Standardized EWW

Time (4 years lag)
Figure 2.18. Measured (solid line) vs modelled (dotted line) earlywood density (EWD), minimum density (MND)
and earlywood width (EWW) for subalpine fir trees at Mt. Chingee and McBride Peak sites along the Rocky
Mountains in northern British Columbia. Data modeled from summer temperature (A), August temperature (B) and
mean previous July and August temperature (C). R2 values are presented with ** P < 0.01 and * P < 0.05. Crossed
solid line indicates 4-year lagged data. The dashed blue trendline represents a 10-year moving trend line.

61

Standardized RW
Standardized MND
Standardized RD
Standardized LWD

Time (4 years lag)
Figure 2.19. Measured (solid line) vs modelled (dotted line) ring width (RW), minimum density (MND), ring
density (RD) and latewood density (LWD) of subalpine fir trees at McBride Peak, Grizzly Den and Averil stands
along the Rocky Mountains in northern British Columbia. Data modeled from previous July temperature (A), April
temperature (B), precipitation of May, June, July and temperature of September, October, November, and December
(C), Precipitation of May, September, October, November and December, and temperature of May and June (D). R2
values are presented with ** P < 0.01 and * P < 0.05. Crossed solid line indicates 4-year lagged data. The dashed
blue trendline represents a 10-year moving trend line.

62

Mt. Averil
RD and LWD chronologies from the Averil site were modelled from a combined precipitation of
May, June, July and temperature of September, October, November, December, and from
Combined precipitation of May, September, October, November and December with temperature
of May and June, respectively (Figure 2.19 C). The RD model corresponded well with combined
precipitation of May, June, July with temperature of September, October, November, December,
and was moderately correlated with measured values (Figure 2.19 C). Modelled RD values
decouple from measured values prior to 1960. However, strong correlations between model and
measure values were obtained from 1965- 1967, 1977-1978 and 2010-2013 (Figure 2.19 C).
LWD chronologies from the Averil site were modelled from combined precipitation of May,
September, October, November and December, with temperature of May and June, and were
moderately related to measured values over most of the time series during 1940- 2018 with some
areas of decoupling occurring in this model between 1969 – 1974 and 1978 - 1990. (Figure 2.19

Standardized MND

D).

Time (4 years lag)
Figure 2.20. Measured (solid line) vs modelled (dotted line) minimum ring density (MND) for subalpine fir trees in
Pine Le Moray site along north-facing slopes of the Rockies in northern British Columbia during 1940 -2018. Data
modeled based on combination of August temperature and September precipitation. R 2 values are presented with **
P <0.01 and * P < 0.05. Crossed solid line indicates 4-year lagged data. The dashed blue trendline represents a 15year moving trend line.

63

Pine Le Moray
MND chronology values from the Pine Le Moray, were moderately related to models based on
mean August temperature and September precipitation (Figures 2.20). Decoupling between
modeled and measured values occurred in 1950-1951, 1963-1964, 1975-1979, 1984-1989 and after
2000 (Figure 2.20). However, strong similarities between modeled and measured MND values
occurred in 1958-1959, 1966-1967 and 1994 (Figure 2.20).

Figure 2.21. Average winter Pacific Decadal Oscillation Index (PDO), and average winter temperatures at Pine Le
Moray, Chingee, Averil, Grizzly Den and McBride sites.

64

Figure 2.22 Average spring Pacific Decadal Oscillation Index (PDO), and average spring temperatures at Pine Le
Moray, Chingee, Averil, Grizzly Den and McBride sites.

2.5 Discussion:
I assumed that subalpine fir growth would be responsive to climate in the locations we targeted
and ring density to be more sensitive to climate than ring width orders. Therefore, according to the
results and existence of significant correlations between subalpine fir radial growth and climate
variables, my hypothesis is confirmed. Due to stronger ring density (MXD, MND, RD)
relationships with climate than annual-ring width (RW) relationships, second part of my
hypothesis is also confirmed.
Most of the study sites showed climatic conditions characteristic of their latitudinal position, and
likely optimal for the support of subalpine fir growth. Pine Le Moray was the highest latitude site
studied and generally had wetter winters, and colder summers compared to other sites (Delong et
al 1993; Wang et al. 2016). The Mt. Chingee, Mt. Averil and Grizzly Den sites occurred at lower
latitudes than Pine Le Moray, but were within 200 m elevation of each other (Table 2.1), and also
experienced wet winters and cool summers. Typically, snowpack at Pine Le Moray remained
65

longer on forest floor than other sites, and comparatively, accumulation of snowpack at Chingee
and Averil sites were higher than at lower latitude sites (Grizzly Den and McBride Peak). Annual
temperature changes at Pine Le Moray, Chingee, Averil and Grizzly sites all show an increasing
trend between 1940 to 2018 while precipitation decreased (Figure 2.2). At the lowest latitude
(McBride site), compared to other sites, climate was drier with an increased air temperature and,
to a lesser extent, increased precipitation between 1940 and 2018 (Figure 2.2).
In general, I came to this results that the optimal growing conditions for subalpine fir growth were
wet climate with storage of water in the soil during dry seasons and the temperature at which trees
showed better signs of optimal growth at study sites (Alexander et al. 1990; Alldritt-McDowell
1998). Because study sites mostly occurred at high latitudes of the Rocky Mountains where
weather was cold, it was not surprising that precipitation was not a limiting factor for growth of
subalpine fir because the sites received enough amount of yearly precipitation (except for some
very dry years that will be discussed later in this section). However, temperature was the more
important climate factor as limiting factor than precipitation for controlling subalpine fir’s growth
at study stands both before and during the growing season considering that most anatomical traits
showed more significant correlations with temperature that with precipitation.
2.5.1 Relationship between climate, radial growth and fibre properties at study sites
Although there were numerous significant correlations between radial growth and climate,
correlations will only be discussed where anatomical wood properties and fibre property models
were successfully predicted from climate variables at statistical strengths of R2 > 0.2 and R2 > 0.7,
respectively. At sites where both FL and MFA, or FWT and MFA, showed significant correlations
with climate, only the strongest correlations will be discussed because fibre length and cell wall
thickness have both been found to be closely correlated with microfibril angle, with smaller angles
66

being related to longer fibre lengths and to thicker walls (Barnett and Bonham 2004; Sheng-Zuo
et al. 2004; Hiller 1964 a,b; Evans et al. 2000). It also has been confirmed that, for a large range
of ring widths, latewood width remains relatively constant between rings, so that variations in ring
width are more closely related to earlywood width (Brazier 1970; Zhang et al. 1996; Savva et al.
2010). Therefore, between EWW and RW properties, I will discuss only those with higher
correlation (i.e. higher R2) with climate. Furthermore, EWD and MND almost tell the same story
and measure almost similar values (not the exact values as EWD is an average measurement of
density across all the earlywood while MND is the measurement of lowest density). I will discuss
only those characteristics that showed statistically stronger and more reliable models for each site
because I found significant overlap when testing both EWD and MND in their relationships to
climate variables.
In the following paragraph I try to meet objective (ii) about evaluating whether intra-annual wood
characteristics show stronger correlations with climate indices or not.
The reasons why intra-annual variations as reflected by monthly climate variables (Figure 2.82.13) showed stronger correlations with monthly climate variables were because in many
applications annual-ring width (RW) estimations give just a growing season portrayal of climatic
information (D’Arrigo et al. 1993). In addition, tree rings for the most part records seasonally
change in climate parameters not at specific and/or short time periods like the beginning or end of
the growing seasons. However, to measure and understand the exact ecological changes even in
small periods of time under changes in climate, intra-annually wood properties estimations
demonstrate higher connections to climatic variables than do RW orders (Wimmer and Grabner
2000; Davi et al. 2002).
I found a lack of significant relationships between climate variables and FWT, MFA and FW at
67

Averil and Grizzly Den, and a lack of significant relationships between climate variables and FL
at Chingee and Pine Le Moray. These findings may have been due to small sample sizes (only 7
to 9 decades were applied using 10-year sections; i.e. n = 7-9) or possibly because of the strength
of influence of genetic factors (Zobel and Buijtenen 1989; Donaldson 2008). Alternatively, other
environmental factors like wind speed, could have influenced MFA development directly as a
result of flexural stress (Wimmer et al. 2002a).
In the following I will provide some evidence to cover objective (i) about comparing variations in
subalpine fir wood properties with climate data:
Influence of temperature
As expected, due to global climate change, mean annual temperatures at all study sites, from lowest
to highest latitudes, have increased by 0.6 ºC to 1.5 ºC during 1940 to 2018 (Figure 2.2). This
increase coordinates with the general increase of 0.9 ºC on average observed across Canada (south
of 60 ºN) over the last century (Zhang et al. 2000). Mean temperature can vary with both latitude
and elevation (Peterson and Peterson 2001). Both the current growing season and pre-growing
season’s temperatures affected wood properties at all five of my study sites. Temperatures within
summer months were by far the most influential climate variables on tree growth and wood
formation in the subalpine fir sampled, according to the relationships identified.
Ring densities generally decreased as summer temperatures increased and summer precipitation
decreased. However, in September, this relationship reversed, and temperature became positively
associated with density at all sites as LW cell needs enough temperature for formation at near the
end of growing season. Normally, cells produced in the early part of the growing season display
larger radial diameter, thinner cell walls with a larger lumen and relatively lower density while
near to the end of growing season, cells with thicker wall and smaller lumen area are produced
68

with higher density (Larson 1969; Worall 1970; Nicholls and Wright 1976; Dutilleul et al. 1998).
Precipitation is the stronger limiting factor for wood density throughout the majority of the
summer, until September when we observe temperature form a stronger relationship to density and
become more limiting (Figures 2.12 and 2.13). This finding is consistent with Wimmer and
Grabner (2000) who showed that average temperature of late summer (September) positively
related to average ring density in Norway spruce (Picea abies).
Increases in air temperature during summer could result in prolonged earlywood formation. At
Pine Le Moray and Chingee, decreases in MND and EWD values during the summer months could
also be associated with higher snowpack accumulation in comparison with cumulative snowpacks
at lower latitude sites. Therefore, larger snowpacks later lead to delay in onset of the growing
season and formation of earlywood cells at Pine Le Moray and Chingee sites. This delay in growth
initiation could mean that production of earlywood cells continued until mid-summer resulting in
lower MND and EWD values at high latitudes. Lesser MND values because of mid-summer
temperature and during the growing season was also found by Franceschini et al. (2013) and was
interpreted as a result of the prolongation of the maturation of earlywood tracheids leading to wider
earlywood cells. Therefore, production of larger rings, led to lower MND values (Mäkinen et al.
2002; Franceschini et al. 2013).
At Grizzly Den, a lower latitude stand compared to the others sampled (Pine Le Moray, Averil and
Chingee), the proportion of earlywood cells increased at the beginning of the growing season
leading to decreased MND values at this site. The accumulation of snowpack was smaller than at
sites located at higher latitudes, caused by increased April temperature which improve thermal
conditions by raising soil temperature in the spring, and removing snow cover allowing trees to
photosynthesize earlier (See Figure A5 and A6 in Appendix) (Tardif et al. 2001a; Rossi et al. 2011;
69

Dolezal et al. 2016).
The transition from higher to lower latitudes along the sampling gradient coordinates with an
amplification of the negative relationship between tree growth and summer temperature. For
instance, at the lowest latitude site (McBride Peak), where weather was warmer and drier and
precipitation was lower than the other sites, previous warmer summers (July and August) (See
Figure A7 in Appendix) could increase water moisture stresses at this site, decrease in
photosynthesis and consequently reduction of growth and EWW values. This result is supported
by King (2013), who suggested that spruce radial growth was increasingly negatively correlated
with previous July and August temperature. Also, Chopin et al. (2002) suggested that prolonged
warm temperatures reduced soil moisture, which was needed for tree growth during low
precipitation periods.
It can be inferred from my results that summer and fall temperature also affected fibre properties
at study sites. Pine Le Moray and McBride show a positive relationship between MFA and summer
temperatures, and Chingee shows the opposite. This indicates that summer temperature, along with
other climate conditions linked to temperature, such as growing season length, were more optimal
for straighter microfibrils at Chingee. Microfibril angle responses to climate is similar to that of
growth rate and ring width (Drew et al. 2013) with the more earlywood cell production associated
with the larger fibre angles (Donaldson 2008). Warmer temperatures in presence of adequate water
and release of water stresses may result in higher MFA, as is supported by Drew et al. (2013).
Conversely higher temperatures and therefore decreased depth and water content of snowpack (or
drought-like conditions) may lead to lower water availability later in the growing season and
reduced MFA and FL. This theory is supported by a study by Jarvis and Linder (2000) who realized
that increased temperatures during fall and winter reduced length of snow cover or winter
70

precipitation depth, or insulation, leading to deeper soil freezing. These dry-like conditions caused
delayed growth onset and shortened growing period which in turn reduced RW and EW cells
(Robertson et al.1990), and resulted in formation of smaller fibre lengths and MFA because both
form at the time of earlywood cells formation (Zhang et al. 2020).
Trees tend to form larger diameter fibres (FW) early in the growing season (Jozsa and Middleton
1994) when earlywood cells are normally produced. Thus, continuous earlywood formation in
summer be a reason for formation of wider fibres at Pine Le Moray. However, at low latitude sites
like McBride, higher than normal temperatures in fall prior to growing season potentially raised
soil temperature leading to increased water stresses (Alfaro 2010) and therefore thinner fibres.
Previous season fall and winter temperatures significantly influenced the growth rates and
formation of annual-rings. Warmer temperatures in seasons prior to growth allow for energy store
to be maximized for use the following year (Fritts 1976; Rolland and Schueller 1994). It is also
possible that the intimate relationship between temperature and snowpacks played a role in the
optimization of growth on sites where snowpack plays an important role in the insulation of roots
during long cold winters, and for the provision of water in spring (Lo et al. 2010; Fleming and
Whitefield 2010). December temperatures influenced fibre length during the following growing
season at Grizzly Den, while fall temperatures and precipitation influenced fibre length in the
following season at McBride; both Grizzly Den and McBride were the lowest latitude sites,
perhaps indicating that lower latitudes rely more heavily on previous years to meet optimal growth
requirements in subsequent years.
Influence of precipitation
Similar to temperature, precipitation also affected wood formation and wood radial growth, but

71

was the limiting factor for growth in fewer instances, as demonstrated by fewer significant
correlations between wood properties and precipitation variables. Since selected study sites were
located at high latitudes and on the north-facing slopes of the Rocky Mountains, the effect of
precipitation on wood properties was less than that of temperature. Spring to early summer
precipitation and precipitation of previous fall affected wood property growth at Averil, Chingee
and McBride sites. According to the results, RD surprisingly increased with spring and early
summer precipitation (May, June and July) at Averil site. Since precipitation falls as snow during
May and early June at high altitude regions in BC (DeLong et al. 1994), increased snow would
lead to a delay in spring snow melt. Therefore, this delay would mean that the growing season
began later in the spring or early summer, and resulted in formation smaller cells and thinner rings
in which proportion of latewood cells were high inducing higher RD at Averil site.
Because McBride Peak was drier than other sites, increased precipitation before the growing
season at this site could have increased soil water storage, and resulted in preservation of enough
water during the growing season necessary for onset and continuation of physiological activities.
Fibre elongation would then be able to start at the first month of the growing season in presence
of optimal moisture and temperature leading to a positive relationship between fibre length and
previous year fall precipitation at the McBride Peak site. Our result is consistent with Around and
Fromm (2003) who suggested that early summer drought significantly reduced fibre elongation.
At higher latitude sites like Chingee, however, the story was different: increased previous fall
precipitation may have increased winter snowpacks, which caused snow cover to stay on the
ground for longer and the growing season to start later. Thus, a shorter growing season resulted in
reduced fibre width.

72

2.5.2. Temporal comparison among sites based on wood fibre property changes under
impacts of microclimate variations and macroclimate thermal phases (PDO)
Over the last 80 years, temperature averages have increased at all study sites while precipitation
decreased at most of the sites (Figure 2.2, 2.3). Temperatures moved from “cool phase” to “warm
phase” (Mantua 2000). Therefore, possibly, a PDO shift from a cool phase to a warm phase may
have influenced the environment for BC’s Forests. It can be understood that likely Pacific Decadal
Oscillation (PDO) spring warm phases in the 1980s and the 1990s played an important role in
increasing temperature at study sites in northern BC and therefore influenced wood fibre property
developments in subalpine fir trees in corresponding sites. This is specifically clear during the
warm period in the 1990s during which subalpine fir trees experienced spring and early summer
drought, resulting in reduced fibre lengths and thicker cell walls (Arend and Fromm 2007).
According to the results, some similar fluctuating trends occurred for FL values at Averil, Grizzly
Den and McBride sites from 1940 to 2018 (Figure 2.14 A; Figure 2.17 A). Precipitation has a basic
and essential role in the formation of tracheid cells in conifers, and FL a basic function to transport
water (Von Arx et al. 2012; Badel et al. 2015), therefore, increasing trends in fibre length were
most possibly due to increases in precipitation from 1940 to 1970 at corresponded sites (Figure
2.3). Following that and during the 1980s and specifically over the 1990s, length of tracheids
decreased at the study sites. Theses deceases in fibre elongation coincided with loss of soil water
due to higher-than-average temperatures in summers during the 1980s and the 1990s at lower
latitude sites (Averil, Grizzly Den and McBride Peak) (Figure 2.2). At a larger scale, fibre length
decreases in corresponded decades could be due to the effects of thermal phases of winter and
spring PDO which reached its peak in the 1990s and influenced local microclimate at study stands
(Figure 2.21; 2.22). In a similar study, Arend and Fromm (2007) reported that early-summer
73

drought significantly reduced fibre length. FW values were also decreased in the 1990s at Pine Le
Moray, Chingee and McBride Peak likely due to higher than average summers in this decade and
lack of soil water storage (Figure 2.14 D; 2.15 C; 2.16 C).
According to the moving averages at Chingee, McBride Peak, Grizzly Den and Pine Le Moray
sites, during 1980s and 1990s minimum density decreased (Figure 2.18 A, B; 2.19 B and 2.20)
which were apparently due to the increased temperature during 1980s and 1990s at corresponded
sites (Figure 2.2) and in a larger scale because of the impact of thermal spring PDO over the whole
region during 1980s and 1990s (Figure 2.22). Because increased temperature during spring induces
earlywood cell productions, MND has been decreased as a result of earlywood cell formation
(Xiang 2014).
2.6. Conclusion
This study aimed to investigate the effect of the most important climate factors including
temperature and precipitation on radial growth properties (RW, EWW, LWW, EWD, LWD, MXD,
MND, RD) and fibre properties (FL, FWT, MFA, FW) of subalpine fir at five study sites along
the north-facing slopes of the Rocky Mountains in northern BC. Annual temperature has increased
at the study sites over the study period since 1940. Annual precipitation trends showed overall
decreases since 1940 at Pine Le Moray and Chingee, while slightly increased at McBride and
remained nearly constant at the Averil and Grizzly sites over past 80 years.
I conclude that relationships between intra-annual-ring characteristics (EWW, EWD, LWW,
LWD, MXD, MND) and climate variables were stronger than simple tree-ring width relationships.
These findings are consistent with other studies that showed intra-annual wood properties
estimations including cell size (cell divider thickness, cell numbers), fibre length, fibre divider

74

thickness and wood thickness demonstrated higher connections to climatic variables than do RW
parameters (Wimmer and Grabner 2000; Davi et al. 2002).
In general, stronger and more robust regression models calculated from the strongest correlations
determined that wood anatomical characteristics of subalpine fir were firstly site specifically
(latitude) affected by climate variables; secondly, strongly influenced by summer temperature,
previous fall temperature, and to a less extent precipitation. The lower the latitude, the greater the
negative effect of summer temperatures on radial growth at study stands.
Late summer precipitation was associated with increased ring density at study stands. It is worth
mentioning that the highest number of anatomical parameters that showed significant relationships
with the climate, associated with the McBride Peak site. This is likely because the McBride Peak
site geographically located at the lowest latitude and warmer temperature and received lower
annual precipitation. Therefore, this site was more sensitive to climate variations than the other
study sites.
Finally, although some wood anatomical characteristics like LWW and RD showed significant
relationships with both temperature and precipitation, most of other properties including RW,
EWW, EWD, LWD, MXD, and MND showed higher correlations with temperature meaning that
temperature was possibly the most important climatic factor controlling growth of subalpine fir at
all five study stands in northern interior BC and along north facing slopes of the Rocky Mountains
since 1940. An old study on subalpine fir in British Columbia also confirmed that temperature was
the most important climate factor affecting growth of subalpine fir (Ettl and Peterson 1995)
perhaps due the position of Canada’s western forests at higher latitudes than the U.S. Furthermore,
in a wider range it has been confirmed that temperature was more important than precipitation
when studying the growth of subalpine trees (Lara et al. 2001; Watson and Luckman 2007).
75

In this study I mostly focused on mathematical and statistical correlations and regression analysis
to find out the relationships between wood characterises and climate indices. Therefore, according
to the significant correlations, it is possible that trees respond to climate factors. However, the
growth is generally based on complex combination of environmental and genetic factors and
therefore, trees are responding to all sort of environmental factors including soil nutrient values,
soil moisture regimes, number of sunny or cloudy days, snowpack accumulations, wind conditions,
stand level conditions and so on. Therefore, I was aware of all sorts of growth-effective parameters
in this study and limitations of this study in calculating subalpine fir responses to climate variables.
Overly, increased awareness of the wood quality (longer fibre, smaller MFA and denser wood) of
subalpine fir under changing climate conditions in this Chapter will feed directly into knowledge
of supply quality for end-uses such as pulp and paper. Results of this study Chapter may also help
to guide timber supply analysis for northern BC. This research could potentially be useful to
improve the Canadian forest industry competitiveness in the global marketplace, in terms of valueadded manufacturing wood industry.

76

2.7 References:
Abatzoglou JT and Williams AP. 2016. Impact of anthropogenic climate change on wildfire across western
US forests. Proceedings of the National Academy of Sciences, 113(42): 11770-11775.
Alexander Robert R. 1987. Ecology, silviculture, and management of the Engelmann spruce-subalpine fir
type in the central and southern Rocky Mountains. USDA Forest Service, Agriculture Handbook 659.
Washington, DC. 144 p.
Alexander RR, Shearer RC, Sheppard WD. 1984. Silvical characteristics of subalpine fir. USDA Forest
Service General Technical Report. RM-115, 60-70.
Alexander RR, Shearer RC and Shepperd WD. 1990. Abies lasiocarpa (Hook.) Nutt. subalpine fir. Silvics
of North America, 1: 60-70.
Alldritt-McDowell J. 1998. Ecology of the Engelmann Spruce-Subalpine Fir Zone.
Alfaro RI, Campbell E and Hawkes BC. 2010. Historical frequency, intensity and extent of mountain pine
beetle disturbance in British Columbia (Vol. 2009). Pacific Forestry Centre.
Arend M and Fromm J. 2007. Seasonal change in the drought response of wood cell development in poplar.
Tree Physiology, 27: 985-992 - doi: 10.1093/treephys/27.7.985
Arend M, and Fromm J. 2003. Ultrastructural changes in cambial cell derivatives during xylem
differentiation in poplar. Plant Biology, 5:255–26
Anderson lV. 1956. Engelmann spruce, its properties and uses. Res. Pap. 39. Ogden, UT: U.S. Department
of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station; 16 p.
Anderegg WRL. 2015. Spatial and temporal variation in plant hydraulic traits and their relevance for
climate change impacts on vegetation. New Phytologist, 205: 1008–1014.
Auty D, Achim A, Macdonald E, Cameron AD, Gardiner BA. 2014. Models for predicting wood density
variation in Scots pine. Forestry, 87: 449-458. - doi: 10.1093/forestry/cpu005
Barnett JR and Jeronimidis (eds). 2003. Tree growth and wood quality. Blackwell Publishing: Oxford (UK).
Badel E, Ewers FW, Cochard H, Telewski FW. 2015. Acclimation of mechanical and hydraulic functions
in trees: impact of the thigmomorphogenetic process. Frontiers in Plant Science, 6 - doi:
10.3389/fpls.2015.00266
Barnett JR and Bonham VA. 2004. Cellulose microfibril angle in the cell wall of wood fibres. Biological
Reviews, 79: 461-472 - doi: 10.1017/s146 4793103006377
Bär A, Pape R, Bräuning A and Löffler J. 2008. Growth-ring variations of dwarf shrubs reflect regional
climate signals in alpine environments rather than topoclimatic differences. Journal of Biogeography, 35:
625-636.
Blanchette D, Fournier RA, Luther JE, Cote JF. 2015. Predicting wood fibre attributes using local-scale
metrics from terrestrial LiDAR data: A case study of Newfoundland conifer species. Forest Ecology and
Management. 347: 116–129 - doi.org/10.1016/j.foreco.2015.03.013

77

Brazier JD. 1970. Timber improvement II. The effect of vigour on young-growth Sitka spruce. Forestry,
43(2): 135-150.
Brooks RJ, Flanagan LB, Ehleringer JR, Responses of boreal conifers to climate fluctuations: indications
from tree-ring widths and carbon isotope analyses. Canadian Journal of Forest Research. 28 (1998): 524533.
Bouriaud O, Bréda N, Dupouey JL, Granier A. 2005. Is ring width a reliable proxy for stem-biomass
increment? A case study in European beech. Canadian Journal of Forest Research, 35(12): 2920–2933doi.org/10.1139/x05-202
Camarero JJ, and Gutierrez E. 2017. Wood density of silver fir reflects drought and cold stress across
climatic and biogeographic gradients. Dendrochronologia, 45: 101-112 - doi: 10.1016/j.dendro.2
017.07.005
Chen HY, and Klinka K. 2000. Height growth models for high-elevation subalpine fir, Engelmann spruce,
and lodgepole pine in British Columbia. Western Journal of Applied Forestry, 15(2): 62-69.
Cregg BM, Dougherty PM, Hennessey TC. 1988. Growth and wood quality of young loblolly pine trees in
relation to stand density and climatic factors. Canadian Journal of Forest Research, 18(7): 851-858.
Cleaveland MK. 1986. Climatic response of densitometric properties in semiarid site tree-rings. Tree-Ring
Bulletin 46:13-29 (includes Corrigendum from Tree-Ring Bulletin 48:41-47.
Cook ER and Peters K. 1981. The smoothing spline: a new approach to standardizing forest interior treering width series for dendroclimatic studies. Tree-Ring Bulletin, 41: 45-55
Cook ER, and Holmes RI. 1999. Users manual for program ARSTAN. Tucson, AZ.
Cochard H, and Telewski FW. 2015. Acclimation of mechanical and hydraulic functions in trees: impact
of the thigmomorphogenetic process. Frontiers in Plant Science 6 - doi: 10.3389/fpls.2015.00266
D'Arrigo RD, Jacoby GC, Free RM. 1992. Tree-ring width and maximum latewood density at the North
American tree line: parameters of climatic change. Canadian Journal of Forest Research, 22(9): 1290-1296.
D'Arrigo RD, and Jacoby GC. 1993. Secular trends in high northern latitude temperature 445
reconstructions based on tree-rings. Climatic Change, 25(2): 163-177.
Davi N, D’Arrigo R, Jacoby G, Buckley B, Kobayashi O. 2002. Warm-season annual to decadal
temperature variability for Hokkaido, Japan, inferred from maximum latewood density (AD 1557–1990)
and ring width data (AD 1532– 1990). Climate Change, 52: 210-217.
Deslauriers A, Morin H, Begin Y. 2003. Cellular phenology of annual-ring formation of Abies balsamea in
the Quebec boreal forest (Canada). Canadian Journal of Forest Research, 33: 190- 200 - doi: 10.1139/x02178.
DeLong C, Tanner D, Jull MJ. 1993. A field guide for site identification and interpretation for the southwest
portion of the Prince George forest region. Province of British Columbia: Ministry of Forests:
https://www.for.gov.bc.ca/hfd/pubs/docs/Lmh/Lmh24.pdf.
Drew DM, Allen K, Downes GM, Evans R, Battaglia M, Baker P. 2013. Wood properties in a long-lived
conifer reveal strong climate signals where ring-width series do not. Tree physiology, 33(1): 37-47.

78

Drew DM, Allen K, Downes GM, Evans R, Battaglia M, Baker P. 2013. Wood properties in a long-lived
conifer reveal strong climate signals where ring-width series do not. Tree Physiology 33: 37-47 - doi:
10.1093/treephys/tps111
Dolezal J, Leheckova E, Sohar K, Dvorsky M, Kopecky M, Chlumska Z, Wild J, Altman J. 2016. Annual
and intra-annual growth dynamics of Myricaria elegans shrubs in arid Himalaya. Trees, 30: 761-773.
Donaldson L. 2008. Microfibril angle: measurement, variation and relationships - a review. Iawa Journal
29: 345-386 - doi: 10.1163/229419 32-90000192.
Downes GM, and Drew DM. 2008. Climate and growth influences on wood formation and utilisation.
Southern Forests: a Journal of Forest Science, 70: 155-167 - doi: 10.2989/south.for.20 08.70.2.11.539
Downes GM, Drew D, Battaglia M, Schulze D. 2009. Measuring and modelling stem growth and wood
formation: an overview. Dendrochronologia, 27: 147-157 - doi: 10.1016/j.dendro. 2009.06.006
Dutilleul P, Herman M, Avella-Shaw T. 1998. Growth rate effects on correlations among ring width, wood
density, and mean tracheid length in Norway spruce (Picea abies), Canada Journal of Forest Research, 28:
56-68.
Ettl GJ, and Peterson DL. 1995. Growth response of subalpine fir (Abies lasiocarpa) to climate in the
Olympic Mountains, Washington, USA. Global Change Biology, 1(3): 213-230.
Evans R, Stringer S, Kibblewhite RP. 2000. Variation of microfibril angle, density and fibre orientation in
twenty-nine Eucalyptus nitens trees. Appita Joiurnal, 53: 450-457.
Fleming SW, and Whitfield PH. 2010. Spatiotemporal mapping of ENSO and PDO surface meteorological
signals in British Columbia, Yukon, and southeast Alaska Spatiotemporal Mapping of ENSO and PDO
Surface Meteorological Signals in British Columbia, Yukon, and Southeast Alaska. Canadian
Meteorological and Oceanographic Society. 48(2): 122-131 - doi.org/10.3137/AO1107.2010
Franceschini T, Longuetaud F, Bontemps JD, Bouriaud O, Caritey BD, Leban JM. 2013. Effect of ring
width, cambial age, and climatic variables on the within-ring wood density profile of Norway spruce Picea
abies (L.) Karst. Trees, 1-13.
Franceschini T, Lundqvist SO, Bontemps JD, Grahn T, Olsson L, Evans R, Leban JM. 2012. Empirical
models for radial and tangential fibre width in tree-rings of Norway spruce in north-western Europe.
Holzforschung, 66(2): 219-230 - doi.org/10.1515/HF.2011.150
Fritts HC. 1976. Tree-rings and Climate. Academic Press, New York.
Fritts HC. 1999. PRECON Version 5.17B. A Statistical Model for Analyzing the Tree- Ring Response to
Variations in Climate. <http://www.ltrr.arizona.edu/~hal/ precon.html>.
Grabner M, Wimmer R, Gierlinger N, Evans R, Downes G. 2005. Heartwood extractives in larch and effects
on X-ray densitometry. Canadian Journal of Forest Research, 35: 2781-2786.
Grissino-Mayer HD. 2001. Evaluating Crossdating Accuracy: A Manual and Tutorial for the Computer
Program COFECHA. Tree-Ring Research, 57(2): 205-221.
Gardiner B, Leban JM, Auty D and Simpson H. 2011. Models for predicting wood density of British-grown
Sitka spruce. Forestry, 84(2): 119-132.

79

Hamann A and Tongli W. 2006. Potential Effects of Climate Change on Ecosystems and Tree Species
Distribution in British Columbia. Ecology, 87(11): 2773-2786.
Haygreen JG and Bowyer JL. 1996. Forest products and wood science: an introduction (No. Ed. 3). Iowa
state university press.
Hiller CH. 1964a. Correlation of fibril angle with wall thickness of tracheids in summerwood of slash and
loblolly pine. Tappi Journal, 47: 125-128.
Hiller CH. 1964b. Estimating size of the fibril angle in latewood tracheids of slash pine. Journal of Forestry,
62: 249-252.
Holmes RL. 1983. Computer assisted quality control in tree-ring dating and measurement. Tree-Ring
Bulletin, 43: 69–78.
Huang CL, Lindstrom H, Nakada R, Ralston J. 2003. Cell wall structure and wood properties determined
by acoustics-a selective review. European Journal of Wood and Wood Products, 61: 321-335 - doi:
10.1007/s00107-003-0398-1
Huang JG, Bergeron Y, Zhai LH, Denneler B. 2011. Variation in intra-annual radial growth (xylem
formation) of Picea mariana (Pinaceae) along a latitudinal gradient in western Quebec, Canada. American
Journal of Botany, 98: 792- 800 - doi: 10.3732/ajb.1000074
IPCC. 2007. Climate Change 2007: the physical science basis. Contribution of working group I to the fourth
assessment report of the intergovernmental panel on climate change. Cambridge University Press,
Cambridge. IUFRO World Series vol. 25, Vienna, Austria, pp. 249-282. [online] URL:
https://www.cabdirect. org/cabdirect/abstract/20113395940
Jacoby GC, and D’Arrigo RD. 1989. Reconstructed northern hemisphere annual temperature since 1671
based on high-latitude tree-ring data from Northern America. Climate Change, 14:39-59
Jacoby GC, and Ulan LD. 1982. Reconstruction of past ice conditions in a Hudson Bay estuary using treerings. Nature (London), 298: 637-639
Jarvis P, and Linder S. 2000. Constraints to growth of boreal forests. Nature, 405: 904-905 doi.org/10.1038/nn.3432
Jensen WB. 2007. The origin of the Soxhlet extractor. Journal of Chemical Education, 84: 1913-1914.
Jiang X, Huang J, Stadt KJ, Comeau PG, Chen HYH. 2016. Spatial climate-dependent growth response of
boreal mixedwood forest in western Canada. Global and Planetary Change, 139: 141-150.
https://doi.org/10.1016/j.gloplacha.2016.02.002
Johns TC, Carnell RE, Crossley JF, Gregory JM, Mitchell JFB, Senior CA, Tett SFB, Wood RA. 1997. The
second Hadley Centre coupled ocean atmosphere GCM: model description, spin up and validation. Climate
Dynamics, 13:103-134.
Jozsa LA. 1991. Wood properties of subalpine fir, a British Columbia Interior wood. BC Ministry of
Forests. FRDA Report, 168, p. 19.
Jozsa LA, Middleton GR .1994. A discussion of wood quality attributes and their practical implications.
Special Publication SP-34, Canada- British Columbia Partnership Agreement on Forest Resource

80

Development - FRDA II, Forintek Canada Corporation, Vancouver, BC, Canada, pp. 42. [online] URL:
http://pdfs.semanticscholar.org/57b5/7386b0f5029e60812695c70 ac34eb624069b.pdf
Keane RE, Mahalovich MF, Bollenbacher BL, Manning ME, Loehman RA, Jain TB, Holsinger LM and
Larson AJ. 2018. Effects of climate change on forest vegetation in the Northern Rockies. In Climate Change
and Rocky Mountain Ecosystems (pp. 59-95). Springer, Cham.
Keane RE, Morgan P, Menakis JP. 1994. Landscape assessment of the decline of whitebark pine (Pinus
albicaulis) in the Bob Marshall Wilderness Complex, Montana, USA. Northwest Science, 68: 213-229.
King G, Fonti P, Nievergelt D, Bu¨ntgen U, Frank D. 2013. Climatic drivers of hourly to yearly tree radius
variations along a 6 _C natural warming gradient. Agriculture and Forest Meteorology, 168:36-46 doi:10.1016/j.agrformet.2012.08.002
Kramer PJ, and Kozlowski TT. 1979. Physiology of woody plants. Academic press, New York.
Klinka K, Wang Q, Kayahara GJ, Carter RE and Blackwell BA. 1992. Light-growth response relationships
in Pacific silver fir (Abies amabilis) and subalpine fir (Abies lasiocarpa). Canadian Journal of
Botany, 70(10): 1919-1930.
Kozlowski TT, Kramer PJ, Pallardy SG. 1991. The physiological ecology of woody plants. Academic Press,
San Diego: 351-371.
Lara A, Aravena JC, Villalba R, Wolodarsky-Franke A, Luckman B, Wilson R. 2001. Dendroclimatology
of high-elevation Nothofagus pumilio forests at their northern distribution limit in the central Andes of
Chile. Canadian Journal of Forest Research, 31:925-936
Larson PR. 1969. Wood formation and the concept of wood quality. New Haven, Yale University.
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.
Lenz O, Schar E, Schweingruber FH. 1976. Methodological problems relative to measurement density and
width of growth rings by X-ray densitogrames of wood. Holzforschung, 30: 114-123.
Lenz P, Cloutier A, MacKay J, Beaulieu J. 2010. Genetic control of wood properties in Picea glauca: an
analysis of trends with cambial age. Canadian Journal of Forest Research, 40: 703-715 - doi: 10.1139/X10014
Liang W, Heinrich I, Helle G, Linan ID, Heinken T. 2013. Applying CLSM to increment core surfaces for
histometric analyses: a novel advance in quantitative wood anatomy. Dendrochronologia, 31:140-145 - doi:
10.1016/j.dendro.2012. 09.002
Lindström H. 1996. Basic density in Norway spruce. Part I. A literature review, Wood and Fibre Science,
28:15-27.
Little RL, Peterson DL, Conquest LL. 1994. Regeneration of subalpine fir (Abies lasiocarpa) following
fire: effects of climate and other factors. Canadian Journal of Forest Research, 24(5):934-944.
Lo YH, Blanco JA, Seely B, Welham C, Kimmins JP. 2010. Relationships between climate and tree radial
growth in interior British Columbia, Canada. Forest Ecology and Management, 259(5): 932-942 doi.org/10.1016/j.foreco.2009.11.033

81

Lokhande S, and Reddy KR. 2014. Quantifying temperature effects on cotton reproductive efficiency and
fibre quality. Agronomy Journal, 106(4):1275-1282
Mantua NJ. 1999: The Pacific Decadal Oscillation. A brief overview for non-specialists, to appear in the
Encyclopedia of Environmental Change.
Mantua N. 2000. The Pacific Decadal Oscillation. Joint Institute for the Study of the Atmosphere and
Ocean. University of Washington. http://jisao.washington.edu/pdo/ Accessed October 2009.
Mansfield SD, Parish R, Di Lucca CM, Goudie J, Kang K-Y, Ott P. 2009. Revisiting the transition between
juvenile and mature wood: a comparison of fibre length, microfibril angle and relative wood density in
lodgepole pine. Holzforschung, 63 - doi: 10.1515/hf.2009.069
Mäkinen H, Nöjd P, Kahle H, Neumann U, others. 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.
Meidinger DJ, Pojar J, editors. 1991. Ecosystems of British Columbia. Research Branch, British Columbia
Ministry of Forests, Victoria, BC. 330 p.
Meko DM and Baisan CH. 2001. Pilot study of latewood-width of conifers as an indicator of variability of
summer rainfall in the North American monsoon region. International Journal of Climatology, 21(6): 697708 - doi.org/10.1002/joc.646
Ministry of Forests and Range. 2008. Emergency Bark Beetle Management Area and Strategic Planning
Maps. Available online://www.for.gov.bc.ca/hfp/mountain_pine_beetle/maps/ebbma/
Morin X, Lechowicz M, Augspurger C, O’Keefe J, Viner D, Chuine I. 2009. Leaf phenology in 22 North
American tree species during the 21st century. Global Change Biology, 15: 961-975 - doi.
org/10.1111/j.1365-2486.2008.01735.x
Mote PW. 2.003. Trends in temperature and precipitation in the Pacific northwest during the twentieth
century. North west Science, 77:271-282.
Nabeshima E, Kubo T, Yasue K, Hiura T, Funada R. 2015. Changes in radial growth of earlywood in
Quercus crispula between 1970 and 2004 reflect climate change. Trees, 29: 1273-1281 - doi:
10.1007/s00468-015-1206-3
Nicholls JWP, Wright JP. 1976. The effect of environmental factors on wood characteristics. 3. The
influence of climate and site on young Pinus radiata material. Canadian Journal of Forest Research. 6: 113121.
Panshin AJ, and deZeeuw C. 1980. Textbook of wood technology. Fourth edition. US: McGraw-Hill Book
Company.
Parmesan C, and Yohe G. 2003. A globally coherent fingerprint of climate change impacts across natural
systems. Nature, 421: 37-42.
Payette S, and Delwaide A. 1994. Growth of black spruce at its northern range limit in arctic Quebec,
Canada. Arctic and Alpine Research, 26: 174-179.
Peterson DW, Peterson DL. 2001. Mountain hemlock growth responds to climatic variability at annual and
decadal time scales. Ecology, 82: 3330-3345.
82

Plomion C, Leprovost G, Stokes A. 2001. Wood formation in trees. Plant Physiology, 127: 1513- 1523 doi: 10.1104/pp.010816
Pitman KJ and Smith DJ. 2013. A dendroclimatic analysis of mountain hemlock (Tsuga mertensiana) ringwidth and maximum density parameters, southern British Columbia Coast Mountains, Canada.
Dendrochronologia, 31(3): 165-174 - doi.org/10.1016/j.dendro.2013.05.002
Pojar J, Klinka, K, Meidinger DV. 1987. Biogeoclimatic ecosystem classification in British
Columbia. Forest Ecology and Management, 22(1-2):119-154.
Rahman H. 2018. Modifying kraft pulping to produce a softwood pulp requiring less energy in tissue paper
production (Doctoral dissertation, Mid Sweden University).
Reddy KR, Davidonis GH, Johnson AS, Vinyard BT. 1999. Temperature regime and carbon dioxide
enrichment alter cotton boll development and fibre properties. Agronomy Journal, 91:851-858 doi:10.2134/agronj1999.915851x
Robertson EO, and Jozsa L. 1990. Estimating Douglas-fir wood production from soil and climate data.
Canadian Journal of Forest Research, 20: 357-364.
Rolland C, and Schueller J. 1994. Relationships between mountain pine and climate in the French Pyreness
(Font-Romeu) studies using the radiodensitometrical method. Pirineos, 43–144, 55–70.
Rossi S. 2003. Intra-annual dynamics of tree-ring formation and effects of the environmental factors at the
timberline in the Eastern Alps. PhD thesis, University of Padova, Padova, Italy
Rossi S, Simard S, Rathgeber C, Deslauriers A, De Zan C. 2009. Effects of a 20-day-long dry period on
cambial and apical meristem growth in Abies balsamea seedlings. Trees-Structure and Function 23:85-93.
Rossi S, Deslauriers A, Anfodillo T, Morin H, Saracino A, Motta R, Borghetti M. 2006. Conifers in cold
environments synchronize maximum growth rate of tree-ring formation with day length. New Phytologist,
170(2): 301–310 - doi.org/10.1111/j.1469-8137.2006.01660.x
Rossi S, Morin H, Deslauriers A, Plourde P-Y. 2011. Predicting xylem phenology in black spruce under
climate warming. Global Change Biology, 17: 614-625.
Rossi S, Girard M-J, Morin H. 2014. Lengthening of the duration of xylogenesis engenders disproportionate
increases in xylem production. Global Change Biology, 20: 2261-2271 - doi.org/10.1111/gcb.12470
Root TL, Price JT, Hall KR, Schneider SH, Rosenzweig C and Pounds JA. 2003. Fingerprints of global
warming on wild animals and plants. Nature, 421: 57-60.
Savva Y, Koubaa A, Tremblay F, Bergeron Y. 2010. Effects of radial growth, tree age, climate, and seed
origin on wood density of diverse Jack pine populations. Trees, 24: 53-65.
Splechtna BE, Dobrys J, 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(2): 89-100.
Stokes MA, and Smiley TL. 1968. An introduction to tree-ring dating. University of Arizona Press: Tucson.
Stocker TF, Qin D, Plattner GK, Tignor MM, Allen SK, Boschung J, Nauels A, Xia Y, Bex V and Midgley
PM. 2014. Climate Change 2013: The physical science basis. contribution of working group I to the fifth
assessment report of IPCC the intergovernmental panel on climate change.

83

Sun Y, Wang LL, Yin H. 2016. Influence of climatic factors on tree-ring maximum latewood density of
Picea schrenkiana in Xinjiang, China. Frontiers of Earth Science, 10: 126-134 - doi: 10.1007/s11707-0150507-6
Sundman O, Reyes DCA, Svedberg A, Hellström S. 2016. On the origin of the spectroscopic signals of
cellulose II-type in sulphite dissolving pulp. Nordic Pulp and Paper Research Journal, 31(1): 54-60.
Sheng-Zuo F, Wen-Zhong Y, Xiang-Xiang F. 2004. Variation of microfibril angle and its correlation to
wood properties in poplars. Journal of Forestry Research, 15 (4): 261-267 - doi: 10.1007/bf02844949
Tardif J, Brisson J, Bergeron Y. 2001a. Dendroclimatic analysis of Acer saccharum, Fagus grandifolia,
and Tsuga canadensis from an old-growth forest, southwestern Quebec. Canadian Journal of Forest
Research, 31: 1491-1501.
Vaganov EA, Hughes MK, Shashkin AV. 2006. Growth Dynamics of Conifer Tree-rings: Images of Past
and Future Environments. Springer-Verlag, Berlin, New York. p 354.
Vieira J, Campelo F, Nabais C. 2009. Age-dependent responses of tree-ring growth and intra-annual density
fluctuations of Pinus pinaster to Mediterranean climate. Trees-Structure and Function, 23: 257- 265.
Von Arx G, Archer SR, Hughes MK. 2012. Long-term functional plasticity in plant hydraulic architecture
in response to supplemental moisture. Annals of Botany, 109: 1091-1100 - doi: 10.1093/aob/mcs030
Van Leeuwen M, Hilker T, Coops NC, Frazer G, Wulder MA, Newnham GJ, Culvenor DS. 2011.
Assessment of standing wood and fibre quality using ground and airborne laser scanning: a review. Forest
Ecology and Management, 261: 1467-1478 - doi: 10.1016/j.foreco.2011.01.032
Watt MS, Clinton PW, Coker G, Davis MR, Simcock R, Parfitt RL, Dando J. 2008. Modelling the influence
of environment and stand characteristics on basic density and modulus of elasticity for young Pinus radiata
and Cupressus lusitanica. Forest Ecology and Management, 255: 1023-1033 - doi:
10.1016/j.foreco.2007.09.086
Wang L, Payette S, Begin Y. 2002. Relationships between anatomical and densitometric characteristics of
black spruce and summer temperature at tree line in northern Quebec. Canadian Journal of Forest Research,
32: 477-486 - doi: 10.1139/x01-208
Wang T, Hamann A, Spittlehouse DL, Murdock TQ. 2012. ClimateWNA-high-resolution spatial climate
data for western North America. Journal of Applied Meteorology and Climatology, 51(1): 16-29.
Wang T, A Hamann, Spittlehouse D, Carroll C. 2016. Locally downscaled and spatially customizable
climate data for historical and future periods for North America. PLoS ONE, 11:1-17.
Walther GR, Post E, Convey P, Menzel A, Parmesan C, Beebee TJ, Fromentin JM, Hoegh-Guldberg O and
Bairlein F. 2002. Ecological responses to recent climate change. Nature, 416: 389-395.
Watson E and Luckman BH. 2006. Long hydroclimate records from tree-rings in western Canada: Potential,
problems and prospects. Canadian Water Resources Journal, 31(4): 205-228.
Watson E and Luckman BH. 2016. An investigation of the snowpack signal in moisture-sensitive trees from
the Southern Canadian Cordillera. Dendrochronologia, 38: 118-130 -doi.org/10.1016/j.dendro.2016.03.008

84

Wimmer R, Downes GM, Evans R, Rasmussen G, French J. 2002b. Direct effects of wood characteristics
on pulp and handsheet properties of Eucalyptus globulus. Holzforschung, 56: 244-252 - doi:
10.1515/HF.2002.040.
Wimmer R, and Grabner M. 2000. A comparison of tree-ring features in Picea abies as correlated with
climate. IAWA Journal, 21: 403-416.
Wimmer R, and Downes GM. 2003. Temporal variation of the ring width-wood density relationship in
Norway spruce grown under two levels of anthropogenic disturbance. IAWA Journal, 24: 53-61.
Worall J. 1970. Interrelationships among some phenological and wood property variables in Norway
spruce, Tappi Journal, 53: 58-63.
Wood LJ, and Smith DJ. 2015. Intra-annual dendroclimatic reconstruction for northern British Columbia,
Canada, using wood properties. Trees, 29: 461-474 - doi.org/10.1007/s00468-014-1124-9
Wood LJ, Smith DJ, Hartley ID. 2016. Predicting softwood quality attributes from climate data in interior
British
Columbia,
Canada.
Forest
Ecology
and
Management,
361:
81-89
doi.org/10.1016/j.foreco.2015.11.004
Wood LJ, and Smith DJ. 2012. Climate and glacier mass balance trends from AD 1780 to present in the
Columbia Mountains, British Columbia, Canada. The Holocene, 23(5): 739–748 doi.org/10.1177/0959683612465450
Xiang W. 2014. Modelling the radial and longitudinal variation in wood density and its sensitivity to
climate in black spruce (Picea mariana (Mill.) BSP) (Doctoral dissertation).
Yamaguchi DK. 1991. A simple method for cross-dating increment cores form living trees. Canadian.
Journal of Forest Research, 21: 414-416.
Zhang S, Belien E, Ren H, Rossi S and Huang JG. 2020. Wood anatomy of boreal species in a warming
world: a review. iForest-Biogeosciences and Forestry, 13(2): 130.
Zhang Q, Alfaro RI, Hebda RJ. 1999. Dendroecological studies of tree growth, climate and spruce beetle
outbreaks in Central British Columbia, Canada. Forest Ecology and Management, 121(3): 215-225.
Zhang Y, Wallace JM, Battisti DS. 1997. ENSO-like interdecadal variability: 1900-1993. Journal of
Climate, 10: 1004-1020.
Zhang SY, Simpson D and Morgenstern EK. 1996. Variation in the relationship of wood density with
growth in 40 black spruce (Picea mariana) families grown in New Brunswick. Wood and Fibre Science,
28: 91–99.
Zhang QB, Hebda RJ, Zhang QJ, Alfaro RI. 2000. Modeling tree-ring growth responses to climatic
variables using artificial neutral networks. Forest Science, 46(2): 229-239.
Zhai L, Bergeron Y, Huang JG, Berninger F. 2012. Variation in intra-annual wood formation, and foliage
and shoot development of three major Canadian boreal tree species. American Journal of Botany, 99: 827837 - doi: 10.3732/ajb.1100235
Zobel BJ, and van Buijtenen JP. 1989. Wood variation and wood properties. In Wood variation Springer,
Berlin, Heidelberg: 1-32.

85

Chapter 3. Legacy effects and growth resilience in northern subalpine fir stands during
severe drought

3.1. Abstract
As global warming continues, drought severity and duration are expected to increase in most areas
especially in dry regions. Therefore, it is very important to understand how trees react to severe
drought periods. In this study I focused on the resilience and recovery of subalpine fir (Abies
lasiocarpa (Hook.) Nutt.)) to severe drought and compared growth resilience and recovery periods
(legacy effects) among five study sites along latitudinal gradient on the north-facing slopes of the
Rocky Mountains in northern BC, Canada. The tree growth recovery period was found to be
between 1 to 2 years for the stands investigated, indicating that subalpine fir trees were resilient
to dry conditions. Dry conditions were defined as those periods with climate moisture deficit
measurements greater than two standard deviations from the average. The lowest latitude stand
had the fastest recover rate from drought. Results also revealed that drought resistance increased
along the latitudinal gradient and toward higher latitudes.

86

3.2. Introduction
Due to global warming, numbers and intensity of drought events may increase in the future (IPCC
2013). The length of drought periods, as well as the severity of drought, are increasing around the
world and consequently, forest dynamics are changing, as forest are vulnerable ecosystems. This
is especially highlighted in mountain regions as it has been shown that temperature is rising faster
than the global mean in mountain ecozones (Gobiet et al. 2014). While some forest ecosystems
benefit from a climate change induced longer growing season (Reyer et al. 2014), other ecosystems
are experiencing drought stresses (Allen et al. 2015). Therefore, depending on the type of forest
and habitat conditions, the behavior of trees in the face of global warming is different around the
world. Meanwhile, forest resilience has become a very important aspect of ecosystem management
and sustainability in addressing climate change and severe climatic events (Seidl 2014; Reyer et
al. 2015). In general, the capacity of a system to rebound or recover from disruption and continue
in the face of perturbations is known as resilience (Carpenter et al. 2001). From an ecological
perspective, resilience is the ability of a system to absorb impacts before a threshold is reached
where the system changes into a different state (Gunderson 2000), while resistance is the capacity
of the ecosystem to absorb disturbances and remain largely unchanged (Holling 1973). It has been
suggested that tree-rings are very informative for studying the effect of drought on forest stands
and forest growth responses to perturbations, as annual tree-rings provide short and long-term
information on how trees respond to drought (Anderegg et al. 2015; Camarero et al. 2015b; Gazol
and Camarero 2016; Gazol et al. 2017c). Studying tree-ring width can help us understand forest
growth responses before, during, and after drought events and provide a measure of resilience of
forests to drought (Lloret et al. 2011).

87

In recent years, dendrochronological methods have been used to study long-term severe climate
impacts on forest ecosystems to assess forest resilience (Lloret et al. 2011; Boden et al. 2014).
Recently, Gazol et al. (2016) studied that, individual tree resilience to perturbance may not be
necessarily representative of stand or landscape level resilience. This is likely why responses of
different tree species to drought and their resilience to severe climate events remains unclear.
Fewer studies have been conducted on tree resilience in dealing with drought (Gazol et al. 2016).
Also, the studies that have been done so far, regarding forest resilience to severe climate using
dendrochronological methods, mostly cover past events and may not necessarily represent future
climate impacts and growth resilience behavior (Radeloff et al. 2015).
Drought indices are one of the most important climatic indicators of resilience in forest stands.
However, not all regions have the same effect (Steinkamp and Hickler 2015). For example, in
boreal forests, reductions in productivity and widespread increases in tree mortality as a
consequence of increasing drought stress have been reported (Peng et al. 2011). Therefore,
predicting the resilience of trees in the face of severe climate events and recovery after drought is
challenging (Gonzalez et al. 2010). Newly, a few studies have suggested that not all regions will
experience widespread forest dieback if severity and duration of drought increase due to climate
change (Steinkamp and Hickler, 2015; Allen et al. 2015). However, it is suggested that even in
cold regions like boreal forests where tree growth is not mainly constraint by drought, reduction
in productivity and widespread increase in tree mortality as a consequence of drought, can be seen
(Peng et al. 2011).
Lately, studies have examined the “legacy effect” on the growth of forest species following
drought (Pretzsch, et al. 2013; Camarero et al 2015; Anderegg et al. 2015; Gazol and Camarero
2016; Wu et at. 2018; Zweifel and Sterck 2018; Huang et al. 2018). Although legacy effect studies

88

of drought are more applicable in arid and semi arid regions (Sánchez-Salguero et al. 2012;
Vicente-Serrano et al., 2013), legacy effects are also studied in some wet and more mesic
ecosystems (Anderegg et al. 2015; Pederson et al. 2014; Peltier, et al. 2016; Gazol and Camarero,
2016; Huang et al. 2018). Legacy effects define as multi‐year lags in tree drought recovery and
can be calculated as the difference between measured and modelled growth (unitless index) after
a severe (2-SD) dry anomaly in the measured climatic moisture deficit (CMD) (Anderegg et al.
2015). Generally, drought happen when precipitation is bellow than evapotranspiration. CMD is
an aridity metric that can be defined in simple terms as the cumulative difference throughout the
year between ET0 (reference evapotranspiration) and precipitation when precipitation is less than
ET0 (i.e., during moisture-stressed months only). Evapotranspiration is the water lost through plant
transpiration and soil and plant evaporation. So, the larger the CMD, the drier condition. CMD has
been used to quantify moisture stress during the dry season which has been shown to drive
vegetation community dynamics in natural ecosystems (Stephenson, 1998) and has been applied
in legacy effect studies (Huang et al. 2018; Kannenberg et al. 2019; Wu et al. 2018). Legacy effects
after drought have also been studied in other plants through stomatal conductance reduction
(Virlouvet and Fromm 2015; Ogle et al. 2015), tree mortality (Anderegg et al. 2012, 2013), xylem
vulnerability (Hacke et al. 2001), and wood anatomy and density (Corcuera et al. 2004). Few
studies have been done to determine how trees use different mechanisms to cope with water deficit
within their wide distributions (Brodribb et al., 2014). Nevertheless, how tree species with different
anatomical, hydraulic and functional traits, respond to drought in terms of growth resilience
remains unclear.
subalpine fir (Abies lasiocarpa) is known as a climate sensitive species in BC, Canada (Zhang
1999). Also, growth of subalpine fir can be restricted by summer temperature specifically on drier

89

sites (Peterson et al. 2002). Warmer summer temperature probably increase summer drought stress
in subalpine fir stands (Peterson et al. 2002) especially on windswept slopes where accumulation
of snowpacks is lighter resulting in soil water stress during summer and lower growth during
drought (Ettl and Peterson 1995). Generally, in subalpine regions, growth condition of trees differ
mostly with latitude and elevation, therefore, growth limiting factors may also change at different
latitudes and elevations probably in response to variation in temperature and soil water availability
(Kienast et al. 1987; Ettl and Peterson 1995). In ESSF zonal climate in BC, drought may occur
due to the interaction of both parameters of lack of rainfall and rising temperatures.
Generally, resilience in subalpine fir stands is component of both individual tree species and
ecosystem. Trees are one component of an ecosystem especially when specific trees are the
dominant plant type in the ecosystem. Therefore, the trees also need to be resilient in an ecosystem.
Individual tree resilience is a reflection of resilience of an ecosystem because trees are indicators
of ecosystem. Resilience of trees to natural disturbances (e.g., severe drought) plays an important
role in ecosystem survival. Trees as the main components of forest ecosystems can maintain the
microclimate conditions inside the forest stands and within an ecosystem. Therefore, other smaller
and understory components of ecosystem can take advantage of dominant tree resilience to
climatic disturbances.
In this chapter, I will describe the ecological resilience of subalpine fir referring to periods when
subalpine fir trees remain dominant and regular in growth pattern compared with period when
subalpine fir shows growth changes due to drought (Gunderson 2000).
3.2.1. Research Question and Objectives
In this study, I address the following question: Do relationships between radial growth and climate
show evidence of resilience to drought, in subalpine fir stands in interior BC? If so, do these
90

relationships show differences in drought resilience with changes in latitude in northern interior
British Columbia? The specific objectives of this research are to:
1) Evaluate patterns in deviations from the average in wood measurements as indicators of
resistance, resilience or recovery after severe climate (e.g., severe drought) in subalpine fir forests.
Compare measurements and trends at five study sites with different latitudes.
2) Evaluate how long (years) it takes to subalpine fir recover after severe drought (legacy
effects) in the five study sites of the Rocky Mountains, British Columbia, Canada.
Hypothesis
Reviewing the literature, I assume that subalpine fir growth resilience and recovery period after
severe drought would vary along latitudinal gradient in response to variation in the dominant
growth limiting factors. Also, I assume that growth resistant to drought at higher latitudes that take
advantage of almost similar colder weather and higher precipitation have similar patterns dealing
with severe drought.
Null hypothesis would be assuming there is no difference in subalpine fir growth resilience by
change in latitude in response to severe drought.
3.2.2. Research Gap
This research will also provide a basis for understanding the resilience of subalpine fir to
accelerated climate change, a knowledge gap that exists due to the lack of information and studies
focus on subalpine fir in the past. Although some legacy effect studies have been started recently
to study growth resilience and recovery period after severe drought around the world (Kannenberg
et al. 2019; Huang et al. 2018; Wu et al. 2018; Peltier et al. 2016; Gazol and Camarero 2016;

91

Anderegg et al. 2015; Pederson et al. 2014), still almost no study has been completed for drought
legacy effects in northern subalpine fir forests in northern BC.
3.3. Methods
3.3.1. Site selection
Subalpine fir trees were selected for core sampling from five natural (not planted or harvested)
stands: Pine Le Moray, Mount Chingee, Mount Averil, Grizzly Den and McBride Peak, all areas
located on north-facing slopes of Rocky Mountains in northern interior BC (Figure 3.1). Distance
between the two farthest sites (Pine Le Moray and McBride Peak) was almost 480 km. The
biogeoclimatic variants of each site were determined with review of site characteristics and
Biogeoclimatic Ecosystem Classification (BEC) land management handbooks, using the BEC
system developed and used within BC to delineate regional differences, topography and dominant
vegetation (Delong et al. 1993; Green et al. 1994) (Table 3.1). Subzones are divisions of zones that
further define the climate of an area. The first part of the subzone name describes the relative
precipitation, and the second part describes the relative temperature (Interior zones). The Pine Le
Moray stand was located within the wet cold variant (wc3) of the Engelmann Spruce-Subalpine
Fir (ESSF) biogeoclimatic zone, classified with high precipitation and cooler temperatures. Mean
annual temperature (MAT) was 1.5 ºC and accumulative annual precipitation (PPT) was 925 mm
with the maximum air temperature occurring in July (19.3 ºC) at this site. The Mount Chingee site
was within the ESSF biogeoclimatic zone, with wet cool variant (wk2). MAT for this site was 0
ºC and PPT was 1060 mm, and maximum air temperature generally occurred in July (16.7 ºC). At
third site, cores were sampled at Mount Averil within the ESSF zone, wet cool variant (wk2),
where MAT was 1.3 ºC and PPT was 1045 mm and maximum air temperature occurred in July
(18 ºC). Grizzly Den, the fourth site, was characterized as ESSF zone, wet cool variant (wk1).
92

MAT for this site was 2 ºC and PPT was 1085 mm and maximum air temperature occurred in July
(19.3 ºC). The last site, McBride Peak, fell within the ESSF moist mild variant (mm1). MAT was
1.6 ºC and PPT was 751 mm at McBride Peak and maximum air temperature occurred in July
(19.6 ºC).

Figure 3.1. Location of sampling sites for legacy effects study in subalpine fir (Abies lasiocarpa) forests along northfacing slopes of the Rocky Mountains in northern British Columbia. From top (highest latitude) to the bottom (lowest
latitude), subalpine fir sampled in Pine Le Moray provincial park, Mount Chingee, Mount Averil, Grizzly Den (Sugar
Bowl) and McBride Peak.

93

3.3.2. Climate Data
The latest dataset of climate information was obtained from modeled climate data; ClimateBC
(http://cfcg.forestry.ubc.ca/projects/climate-data/climatebcwna/) for the five study stands for the
period between 1940 to 2018. ClimateBC provides climatic informative models for each region by
entering latitude, longitude, and altitude of target site. The climate variable representative of
drought events in ClimateBC is the climate moisture deficit (CMD) which is a common variable
in legacy effect studies (Anderegg et al. 2015; Wu et al. 2018; Kannenberg et al. 2019). CMD is
calculated from the equation of precipitation and evapotranspiration. Evapotranspiration increases
with air temperature, solar radiation, and wind speed (Allen et al. 1998). Severe drought events
are represented by a twofold standard deviation dry anomaly (i.e., 2 SD) of synthetic drought index
or calculated Climatic Moisture Deficit (CMD) variables.
3.3.3. Sample Collection
Twenty-two dominant subalpine fir trees at each stand were selected for sampling. To avoid spatial
autocorrelation trees were sampled at 5m minimum intervals (Dale and Fortin 2014). Healthy trees
(no scratch, fire or insect damage or abnormal growth) with full crowns were targeted. Two 12
mm in diameter cores, perpendicular to the axis of each tree, were collected at breast height (1.3
m height). Increment cores were sampled parallel to the topographic contour lines of the slope to
avoid the effect of reaction wood (Pourtahmasi et al. 2011). Surrounding vegetation, slope,
elevation, diameter-at-breast-height (measured in the lab by adding two radial core samples
together per tree) were recorded (Table 3.1).

94

Table 3.1. Sampling site collection information and individual chronology characteristics for five study stands along
the Rock Mountains in northern BC. Biogeoclimatic variant of each site was determined with review of site
characteristics and biogeoclimatic ecosystem classification land management handbooks (DeLong et al. 1994).

Geographic information
Site

Latitude Longitude Aspect

Elevation
(m)

BEC
zone

Tree characteristics
BEC
Mean
Mean age # of tree
subzon
DBH
(years)
cores
e
(cm)

Pine Le Moray

55° 24ʹ

-122° 32ʹ

N

1120

ESSF

wc3

90

22

43

Mt. Chingee

55° 01ʹ

-122° 53ʹ

N

1200

ESSF

wk2

76

22

43

Mt. Averil

54° 24ʹ

-122° 25ʹ

NW

1270

ESSF

wk2

104

22

36

Grizzly Den

53° 46ʹ

-121° 30ʹ

N

1250

ESSF

wk1

107

22

42

McBride Peak

53° 19ʹ

-120° 07ʹ

W

1330

ESSF

mm1

71

22

41

3.3.4. Sample Preparation of 12mm Cores
Of the 110-12mm diameter cores sampled in this study from five study sites, 93 undamaged (not
broken) cores were selected for the legacy effects study. Cores were extracted using 12 mm
diameter increment borer and transferred to the UNBC Dendrochronology Lab. To dry the
samples, cores were submerged into 95% ethanol for 48 hours. This process helps water molecules
inside core samples be replaced with alcohol molecules to dry faster. Once all liquids were
evaporated, cores were mounted on grooved boards and were cut into 2 mm radial pith-to-bark
strips (tangentially x longitudinally) using a twin-blade saw. Resins were then removed from 2mm
laths using Soxhlet acetone extraction for 5 hours. After extracting chemical compounds, samples
were scanned using ITRAX densitometers, which use an X-ray laser beam to provide
measurements of wood density (Evans and Ilic 2001). Radiographs obtained from software
connected to the ITRAX machine were transferred into WinDendro for measuring ring width and
earlywood width.

95

3.3.5. Master chronology
Radiographic core images were then crossdated in two steps; 1) Chronologies were visually
crossdated using the skeleton plot method in WinDendro by identifying specific years (especially
narrow rings) as significant marker years (Yamaguchi 1991); 2) Visually crossdated chronologies
were then verified using COFECHA (Grissino-Mayer 2001). Laths were labelled with a single dot
every ten years, two dots every half-century and three dots every century (Stokes and Smiley
1968). Wood laths with unique variations not representative of the stand that could not be matched
with other samples within a site, were removed from further analysis. By comparing each core with
the mean chronology of each site, COFECHA program generates the mean correlation between
the series (MSI). COFECHA also takes note of individual series that may contain errors such as
missed or false rings. The measurements are then validated until the remaining indicators
disappear or until the MSI is well above 0.42, the minimum value of significance at the 99%
confidence level (Grissino-Mayer, 2001). Once crossdated, step standardization was conducted
using ARSTAN software and negative exponential curving method to remove non-climatic factors
affecting the growth from the series resulting in dimensionless (index) values (Cook and Holmes
1999). In case first detrending curve did not fit to the series, I applied Hugershoff curve to match
the chronologies and detrend the series. ARSTAN removes all non-climatic trends from series
including the most important one, age related variations, knowing that young trees have faster
growth in comparison to mature trees (Cook and Peters 1981).
Each series was transformed and averaged to create a master site chronology for all wood
characteristics using running window of 50 years with an overlap of 25 years for sites with at least
hundred years of each chronology (Pine Le Moray, Chingee and Grizzly Den), and running
window of 40 or 30 years with overlap of 20 or 15 years for shorter (Averil and McBride)

96

chronologies (Cook and Holmes 1999). EPS (expressed population signal) was used (Table 3.2)
to determine when chronologies had low sample size near the beginning of a series and the
accuracy (EPS >85%) of sample representation of the population (Wigley et al. 1984). Study site’s
mean chronology lengths were between 67-101 years (Table 3.2). The years of EPS cut-off were
varied between 1940 - 1960 in study stands (Table 3.2). Stabilization of RW and EWW was
achieved through application of a negative exponential curve (one-time detrending) or straightline fit followed by a smoothing spline in ARSTAN (Cook and Peters 1981; Meko and Baisan
2001). Standard series obtained from ARSTAN at each site and were used for comparison with
climate indices.
Because changes in the size of rings width (RW) or earlywood width (EWW) show the effect of
drought better and more understandable than ring density, I used these traits for the legacy effects
study. Chronologies from each stand were tested for normality using skewness and kurtosis values
prior to correlating with CMD (Table3.3). Normality testing of RW and EWW chronologies
determined most data as normal except for Averil-RW and EWW (Table 3.3). Since all data must
be normal for legacy effect studies (Wu et al. 2018), non-normal data were also normalized. The
formula that we used to normalize non-normal data value (e.g., x), was as follows:
Normalized value = (x – x̄) / s
where:
x = data value
x̄ = mean of dataset
s = standard deviation of dataset

97

Table 3.2. Master chronology statistics for five stands within central interior British Columbia. Statistics shown
include chronology length, mean sensitivity and EPS (expressed population signal) cut off and number of trees
successfully crossdated and standardized for tree-ring width (RW), and earlywood width (EWW). MSI = Mean series
intercorreltion: indicates the strength of correlation between samples taken from the same location, based on 50-year
overlapping segments

Site Name

Chronology
Type

Pine Le
Moray

RW
EWW

0.563

Mt.Chingee

RW
EWW

0.664

Mt. Averil

RW
EWW

0.744

RW
EWW

0.693

RW
EWW

0.743

Grizzly
Den
McBride
Peak

MSI

0.612
0.596
0.696
0.722
0.777

EPS
Chronology
cutoff
Length

Mean
EPS
Length
value of Series
(year)

Accepted
series
Length

# of Trees
# of Trees
Successfully Successfully
Crossdated Standardized

1915-2018 1940 >84%

85

80

19

16

1923-2018 1950 >83%

67

70

18

15

1880-2018 1940 >84%

101

80

19

16

1870-2018 1940 >83%

101

80

18

15

1932-2018 1960 >83%

64

60

18

15

Table 3.3. Normality statistics (Skewness and Kurtosis) for ring width and earlywood width chronologies measured
at Pine Le Moray, Chingee, Averil, Grizzly and McBride sites along the Rocky Mountain in northern BC. Data
normality tests for Skewness and Kurtosis z-values > +1.96 or < -1.96 were not accepted as normal data. Bold values
indicate non-normal values. SE= standard error.

Tree-Ring Width
Site
Pine Le Moray
Mt. Chingee
Mt. Averil
Grizzly Den
McBride Peak

Z-value
Skewness (dividing
skewness
SE
by its SE)

0.020
0.290
0.007
0.310
0.840
0.270
-0.150
0.270
-0.120
0.310

0.069
0.023
3.111
-0.556
-0.387

Kurtosis
SE

-0.440
0.570
-1.170
0.610
0.640
0.530
0.440
0.530
-0.590
0.610

Earlywood Width

Z-value
Z-value
Skewness (dividing
(dividing
kurtosis by
skewness by
SE
its SE)
its SE)

-0.772
-1.918
1.208
0.830
-0.967

98

0.110
0.290
-0.010
0.310
0.760
0.270
0.080
0.270
0.210
0.310

0.379
-0.032
2.815
0.296
0.677

Kurtosis
SE

-0.460
0.570
-1.170
0.610
0.590
0.530
-0.950
0.530
0.770
0.610

Z-value
(dividing
kurtosis by
its SE)

-0.807
-1.918
1.113
-1.792
1.262

3.3.6. Resistance, Recovery and Resilience
To study the growth of subalpine fir trees under episodic drought stress, I applied indices for
resistance (Rt), recovery (Rc) and resilience (Rs) as introduced and explained in detail by Lloret
et al. (2011) and Pretzsch et al. (2013). The three indices were calculated for RW or EWW
(standardized indices) using the master chronologies for each site. Basic components of all three
indices are as follow: Gpre is 2-year mean ring width growth (RW) or earlywood width growth
(EWW) in a period of the nPreDr years before the drought period; Gd is the mean RW or EWW
in a period of nDr years during the drought period; and Gpost is the mean RW or EWW in a period
of nPostDr years after the drought period. The resistance, Rt = Gd ⁄ Gpre, quantifies the decrease
in growth from the pre-drought period to the drought period; Rt = 1 is a complete resistance (no
change in growth); the further the value falls below Rt = 1, the lower the resistance (Figure 3.2).
Recovery, Rc = Gpost ⁄ Gd, describes the growth reaction after the drought period. Rc = 1 indicates
persistence of a low growth level even after 1-year length of the drought period (nDr = 1)
(assuming an Rt < 1 or no change in growth if Rt = 1). Rc >1 represent fully recovered growth
after one-year severe drought (Figure 3.2). Resilience, Rs = Gpost / Gpre, measures the difference
in level of PreDr and PostDr. Rs < 1 shows low resilience of subalpine fir trees after drought
condition. Rs = 1 and Rs > 1 represent good resilience and fully resilience of growth after severe
drought, respectively (Figure 3.2).

99

Figure 3.2. Course of growth, ring width (RW) and earlywood width (EWW), in three different stress events characterised by growth in the period before drought
(PreDr), growth in the drought period (Dr) and growth after the drought period (PostDr). Indices for resistance; Rt = Dr ⁄ PreDr, recovery; Rc = PostDr ⁄ Dr, and
resilience, Rs = PostDr ⁄ PreDr, are used to characterize the stress response patterns. (A) Tree with growth decrease by drought (Rs= PostDr ⁄ PreDr) if Rs< 1
indicated by low resilience (assuming drought always causes an initial decline in growth). (B) Tree with growth recovered after drought (Rs= PostDr ⁄ PreDr) if
Rs= 1 indicated by complete resistance while Rt <1 indicated by weak resistance. (C) Tree with growth considerably recovered after drought (Rs = PostDr ⁄ PreDr)
if Rs> 1 indicated by high resistance. In the graphs Rt, Rc and Rs are represented by the gradient of decline from PostDr to Dr, the increase from Dr to PostDr, and
the difference in level of PreDr and PostDr, respectively (Presented in figure 1 by Lliot et al. 2011 and figure 1 by Pretzsch et al. 2013).

100

3.3.7. Legacy Effects
I defined severe drought legacy in tree growth as a departure of the observed unitless growth (RW
or EWW) from predicted tree growth in a period ranging 1-4 years after a severe drought event for
each of study stands following the methods Anderegg et al. (2015). Only single drought events (no
consecutive drought within 4 years after a -2 SD dry anomaly) lasting no more than 1 year were
considered because capacity of trees for recovery after successive droughts reduce considerably
(Peltier et al. 2016). The “predicted” growth in 1-4 years after severe drought event was calculated
using linear regressions over the entire period by comparison between measured RW or EWW and
the drought variable (Climate Moisture Deficit). Informative prediction of legacy response of
vegetation growth after severe drought events depends on the strength (i.e., the goodness of fit) of
the linear regressions between tree growth and the drought variable.
3.3.8. Analysis and Comparison of Measured and Modelled Values
Regression analysis was completed where strong, significant Pearson’s or Spearman’s correlation
coefficients were identified, above 0.3 (Anderegg et al. 2015) (Table 3.4). Regression models with
significant correlation coefficients were visually assessed against measured values to determine
model accuracy over time. To obtain stronger models, if needed we combined two climatic
variables (CMD) in the same model to predict better and stronger models provided that the
independent variables (climatic parameters) were not statistically correlated. To find if variables
are related together or not, a Durbin-Watson statistic was used. If independent variables were
autocorrelated we could potentially take an average of two independent variables and put it in the
model for prediction. Split verification was then used to validate the models. The latter 50% of
time series was applied to calibrate the models. All predicated site-specific anatomical traits passed
the verification test at 90% significance of reduction of error values (RE) compared to actual
101

measured data.
3.4. Results
Generally, since there are significant correlations (P < 0.05) between tree-ring width chronologies
and severe drought, null hypothesis is rejected.
3.4.1. Correlation and regression analysis for growth models
Durbin Watson test results showed that spring and summer climate moisture deficits (CMD) were
related (Appendix A). Therefore, where needed, a measured average of spring and summer CMD
was used to obtain stronger models for Averil and McBride sites (Table 3.4). Because the objective
of this study was to calculate and to measure duration of growth suppression after drought
episodes, I accepted only severe drought event to growth (RW/EWW) significant correlations
where r ≥ 0.3 (Anderegg et al. 2015). The models presented reflect the strongest relationships
found between subalpine fir growth and climate moisture deficit (Figure 3.4). Since no significant
correlation was found between subalpine growth (RW/ EWW) and CMD at the Grizzly Den site,
regression analysis could not be completed for this site and therefore the Grizzly Den stand was
removed from further legacy effect measurement in this study. However, significant correlations
were found from comparison between tree-ring width or earlywood width and severe drought
events (>2 SD from climatic moisture deficit) that lasted 1 to 2 years after drought event in the
other four study stands including Pine Le Moray, Chingee, McBride and Averil (Table 3.4). Since
correlation between RW and EWW (Table 3.4) was high, RW and EWW were replaced where one
showed higher correlation with drought.

102

Table 3.4 Pearson correlation coefficient analysis between growth representatives (RW/EWW) and CMD, and
between Ring width (RW) and earlywood width (EWW) of subalpine fir in all five study stands along the north-facing
slopes of the Rocky Mountains. CMD= mean climate moisture deficit, sm= Summer, sp= Spring, *= significant
correlations at 95% confidence level, **= significant correlations at 99% confidence level. Bold figures represent r ≥
⁺₋ 0.3 which were accepted for legacy effect study.

EWW

RW

Chingee

Grizzly

Averil

McBride

Pine

RW

0.959

0.8

0.961

1

1

CMD

-0.365**

0.068

-0.187

-0.185

-0.131

CMD (sm,sp)

0.04

-0.095

-0.3*

-0.4**

-0.189

CMD (sm)

-0.257

0.127

-0.139

-0.136

-0.31*

103

104

Figure 3.3 Climate Moisture Deficit (CMD) during 1950-2018 in four study regions around PG and along north facing
slopes of the Rocky Mountains in northern BC. (A) CMD recognised as drought (> 1-SD) and severe drought (>2SD) at the Mount Chingee site where three years (1967, 1985 and 1992) are marked as severely dry years measuring
2-SD mean CMD during 1961-2018. (B) Combined mean summer and spring CMD were used to calculate dry years
and very dry years (1960 and 1992) at the McBride stand measuring 2-SD during 1957-2018. (C) Mean summer CMD
was applied to measure dry and severely dry years (1967, 1992 and 2006) at the Pine Le Moray stand during 19562018. (D) Combined mean summer and spring CMD were used to extract very dry years (1958, 1967, 1985 and 1992)
during 1950-2018 at the Averil site. Gray dotted lines represent the lowest levels at which 1-SD and 2-SD were met.
Small circle points show CMD> SD. Larger gray circles represent severe drought years for which CMD > 2-SD. SD
= standard deviation, CMD = climate moisture deficit.

105

Table 3.5. Resistance, recovery, and resilience indices of Subalpine fir after sever drought and growth index prior, during and after severe drought events (> 2-SD)
in Chingee, McBride, Pine Le Moray and Averil stands around PG and along north-facing slopes of Rocky Mountains during 1960-2015. Gpre = growth prior to
severe drought events, Gpost = Growth after severe droughts, Gd = growth during severe drought, Rt = resistance, Rc = recovery and Rs = resilience.

Site

Pine Le Moray

Mt. Chingee

Mt. Averil

McBride Peak

Severe Drought
year

Gpre
(2 yeras)

Gpost
(2 years)

Gd

1967

1.025

0.79

0.92

1992

0.95

1.08

1.19

2006

0.93

0.91

1.02

1967

1.02

0.94

1.001

1985

1.04

0.93

0.907

1992

0.86

0.87

1.089

1958

0.3

0.09

0.24

1967

0.3

0.09

0.2

1985

0.31

0.23

0.33

1992

0.32

0.35

0.61

1960

1

0.96

0.72

1992

1.1

1.13

0.94

106

Mean Rt

Mean Rc

Mean Rs

1.08

0.88

0.96

1.04

0.92

0.94

1.10

0.52

0.60

0.82

1.26

0.99

3.4.2. Severe drought, legacy effect, and resilience studies in subalpine fir stands
My results showed that recovery of subalpine fir growth after a severe drought event (defined as a
year when the CMD was greater than two standard deviations from the mean), can take from one
to two years (Figure 3.4). Three years (1967, 1985 and 1992) were recognized and calculated as
severely dry years at Mount Chingee, two years (1960, 1992) at McBride Peak, three years (1967,
1992, 2006) at Pine Le Moray and four years (1958, 1967, 1985, 1992) at Averil site by measuring
2-SD mean CMD (Figure 3.3). For the reference periods before and after the dry years, we finally
used a 2-year length (nPreDr = 2, nPostDr = 2). Although growth periods during very dry years at
study sites lied almost within middle-aged phase of chronology length and stress response patterns
may change with tree age and tree size (Carrer and Urbinati 2004; Rozas 2005; Rossi et al. 2008),
standardized growth (transformation by detrending and indexing of the growth rates) for both
EWW and RW was applied in this study to remove all age-related trends.
The average growth before, during and after very dry years was noted at study sites. Rt, Rc and Rs
were calculated for the Chingee stand as 1.04, 0.92 and 0.94 respectively (Table 3.5). Mean
measured growth prior to, during and after severe drought was used to calculate Rt (0.82), Rc (1.26)
and Rs (0.99) for the McBride stand (Table 3.5). The third site, Pine Le Moray, Rt (1.08), Rc (0.88)
and Rs (0.96) were calculated from mean growth before, during and after severe drought (Table
3.5). The fourth site, Averil, Rt (1.1), Rc (0.51) and Rs (0.6) obtained from growth in conditions
prior, during and after very dry conditions at this site (Table 3.5). According to the results (Table
3.5, Figure 3.4), resistance of subalpine fir growth to drought increased with latitude and
precipitation, whereas recovery of tree growth after drought decreased with latitude. The analyses
revealed that the growth resilience index (Rs) of subalpine fir varied across the studied stands
(Table 3.5).

107

Drought legacy effects lasted between 1 and 2 years at the study sites. Results for the Mount
Chingee stand showed severe dry years after which departure of observed growth level was less
than average predicted growth that lasted 1 year after severe droughts (Figure 3.4 A). These effects
were substantial in magnitude: there was an average of a ~13% decrease in observed versus
predicted growth in year 1 after severe droughts in Chingee stand. The McBride, very dry years
determining significant departure of observed growth versus predicted growth where legacy effects
lasted 1 year after severe drought conditions (Figure 3.4 B). The magnitude of the legacy effect
was an average of a 14% decrease in measured growth than predicted growth for the McBride
stand. Substantial severe dry years were recognized as significant departures between measured
and predicted subalpine fir growth at the Pine Le Moray stand during 1956- 2018 (Figure 3.4 C).
Magnitude of these departures were a 20% decrease in observed vs predicted values at year 1, and
17% decrease at year 2 after drought, average, at the Pine Le Moray stand (Figure 3.4 C). For
Averil, specific drought years with significant departure of measured growth from predicted
growth were observed (Figure 3.3 D). Differences between observed values and predicted values
showed a magnitude of a 22% decrease in average observed growth versus predicted growth in
year 1 and 33% decrease at year 2 after very dry conditions at Averil site (Figure 3.4 D). Our
analysis revealed that recovery (Rc = 1.26) and resilience (Rs = 0.99) of subalpine fir at the McBride
site were greater than other three stands (Table 3.4). However, growth resistance to severe droughts
at Averil (Rt = 1.1), Pine (Rt = 1.08) and Chingee (Rt = 1.04) sites were higher than the McBride
(Rt = 0.86) site. It is interesting to note that growth recovery at McBride Peak site was higher than
1 (Rc = 1.26) and growth resilience was very close to 1 (Rs = 0.99) meaning that after severe
drought years at this site, growth was fully recovered. Also, in comparison to the other sites, the
McBride site was more resilient to very dry conditions (Table 3.5; Figure 3.4).

108

Figure 3.4. Drought legacy during 1–4 years post severe drought years obtained from differences between observed values and predicted values that extracted from
correlations between Ring-Width Index (RWI) or early wood width index (EWW) and climatic moisture deficit (CMD) in four study regions around PG and along
the Rocky Mountains in northern BC; (A) legacy effect in Chingee site that longed 1 year after mean sever drought between 1960-2018, (B) legacy effect in
McBride site that lasted 1 year after mean sever drought from 1960 to 2018, (C) legacy effect in Pine Le Moray site that longed 2 years after mean sever drought
between 1950- 2018, and (D) legacy effect in Averil stand that continued 2 years after mean sever drought from 1940 to 2018. Dashed lines mark the level where
drought happens.
= Onset of drought event

109

3.5. Discussion
Study of legacy effects in this research was done for the only single drought events (defined as, no
consecutive drought within 4 years after a -2 SD dry anomaly) lasting no more than 1 year because
the capacity of trees for recovery after successive droughts reduce considerably (Peltier et al.
2016). It also has been shown that if the time between drought events becomes shorter than the
time needed for trees to recover from them, trees are likely to become increasingly prone to water
stress and insect attacks (Schwalm et al. 2017).
Grizzly Den site was not used for further analysis because there was no correlation between
chronologies and drought. This indicates that the trees at Grizzly Den site were responding to
climate factors outside of what I investigated. Therefore, this site has other variables that I did not
measure that could be influencing growth more than what I did measure in this research.
Different growth resilience and recovery period (the length of time that a tree can compensate for
the decline due to drought) in subalpine fir stands by comparing growth patterns after drought at
higher latitude sites and lower latitude site, consists with my hypothesis and strengthen it. Similar
resistance to drought at higher latitudes which may be due to the colder temperature and
accumulation of snowpack also consist with my second hypothesis. This high depth of snowpack
at higher latitudes could likely persist long enough till first months of the summer to keep soil
water balance.
Drought induced a decline in subalpine fir growth in our study. This pattern is not unexpected, as
drought-driven declines in forest productivity and widespread increases in tree mortality have been
observed at the global scale and across different biomes (Allen et al. 2010), a trend that is expected
to continue into the future as climates warm (Schwalm et al. 2017). Also, this finding is consistent
with a study by Peterson et al (2002) who revealed that subalpine fir growth declined with warmer
summer in the western north US.
110

Although this study has been conducted in mountainous and cold regions of the Rockies and may
not be typical place for legacy
effect studies, recent investigations of cold and more mesic forests have highlighted their
susceptibility to current and future water stress at these regions (Allen et al. 2010; Pederson et al.
2014; Martin-Benito and Pederson 2015; Millar and Stephenson 2015).
I observed legacy effects on radial growth after severe drought at all sites and negative effects of
drought on ring width. These results are consistent with more recent studies about legacy effects
that have shown similar negative correlations that exist between drought and tree radial growth
(Anderegg et al. 2015; Camarero et al. 2015; Gazol and Camarero 2016; Wu et at. 2018; Zweifel
and Sterck 2018).
In the following I will provide some evidence for addressing objective (i) and (iii) about existence
of resistant, recovery and resilience at subalpine fir stands, and difference in subalpine fir resilience
at difference sites, respectively:
It can be inferred from results that recovery and resilience of subalpine fir trees to severe droughts
at high-latitudes (Pine Le Moray, Chingee and Averil) stands were lower than the stand located at
the lowest latitude (McBride Peak). Conversely, growth resistance of subalpine fir trees after
severely dry years at high-latitude (Pine le Moray, Chingee and Averil) stands were greater than
lower latitude stand (McBride Peak) site. These results are consistent with Gazol et al. (2016) who
showed forest growth displayed different resilience at different latitudes, and while drought
resistance increased along gradients of latitude and longitudinal as well as soil moisture, drought
recovery decreased. Ettl and Peterson (1995) also found that growth response of subalpine fir to
climate varied between wet and dry sites which was likely due to different stands soil water
availability at different sites.

111

Higher resistance to drought at higher latitudes in comparison to lower latitude sites was due to
colder weather and deeper snowpacks in winter and spring that persist long enough till first months
of the summer when drought mostly happens. Therefore, soil moisture is not as limiting to growth
on these sites and summer drought stress reduce by reduction evapotranspiration rates at higher
latitude subalpine fir sites (Peterson et al. 2002).
Based on my results, legacy effects were detected within a maximum time lag of one year after
severe drought at the McBride Peak and Chingee sites. These results show that recovery after
drought was high at McBride Peak but to a lesser extent than at the Chingee site (Figure 3.4; Table
3.5). However, resistance of subalpine fir trees at McBride was lower than other study regions.
Inverse relation between resistance response and resilience to drought can be supported by other
studies that showed there is a negative relationship between resistance of forest growth to drought
and forest recovery from drought (Lliot et at. 2011; Prestch et al. 2013). Our finding that fast
recovery of growth after drought (Anderegg et al. 2015) in lower latitude forests (McBride Peak
site) is supported by Beck et al. (2011) who suggested that trees at warmer sites may take advantage
of favourable longer growing season conditions compared to colder regions. There are many
reasons why the recovery period in a forest stand may be higher than another stand. For instance,
variability in soil moisture is an important factor affecting length of recovery after drought. At
lower latitude, winter snowpack does not stay for a long time on the forest floor in spring, and may
melt faster than high latitude regions. Therefore, at lower latitudes optimum growth condition
(optimal temperature and availability of water) is met earlier in spring and growth period extend,
therefore trees have more time to compensate for lower growth of the drought. Higher temperature
at lower latitudes could significantly reduce snowpack depth in winter and increase precipitation
fall as rain (Mote 1999). This condition reduces snowpack accumulation in spring resulted in

112

earlier initiation of growing period and let trees to compensate lower growth of previous year,
faster.
At higher latitudes, conversely, accumulation of snowpack and colder weather conditions may
result in delayed snowpack melting on forest floor, especially during the first months of the
growing season, leading to dry-like condition on forest floor. As a consequence, probably shorter
growing periods in current year can not compensate lower growth of previous year (drought year).
This idea and my results are in consistency with other researchers who suggested that higher
latitudes in the (sub) alpine zone are likely to constrain fast recovery after drought (De Boeck et
al. 2016; Cremonese et al. 2017). Peterson et al. (2002) also found that depth of spring snowpack
in subalpine fir stands persist long at higher latitudes and colder sites and produced a growing
season length limitation specially on northern slopes.
In the following paragraphs I will provide some evidence to cover objective (i) about duration of
drought legacy effect at study stands:
Anderreg et al. (2015) suggested that legacy effects in boreal forest usually takes 2-4 years. In our
study, legacy effect in subalpine fir trees from four study sites along the Rocky Mountains took
between one to two years after drought. I found that drought legacy effects in the study regions
did not necessarily cause weaker drought resilience (i.e., much larger reduction in forest growth
after drought). Instead, I observed drought resiliency (i.e., smaller growth reduction after severe
drought) (Gazol et al. 2017a) in study stands. This 1-to-2-year(s) recovery after drought
(resilience) may be due to the specie’s hydraulic conductivity in drought condition, which is related
to xylem plasticity, which may substantially help to increase resilience, and tolerate very dry
conditions (Lopez-Iglesias et al. 2014; Anderegg et al. 2015; Anderegg and HilleRisLambers
2016; Lopez et al. 2016). Apparently due to having some morphological changes in response to

113

dry condition (Kuuluvainen et al. 1996), subalpine fir is considered as a drought tolerant species
(Hinckley et al. 1982). One thing that should be considered is that differences in growth responses
to drought among sites may depend on difference in drought intensity among study sites (Zhang
et al. 2017) or specific functional traits (Greenwood et al. 2017) or individual specific strategies to
cope with drought (Gazol et al. 2017).
In this study, I found that legacy effects reduced growth of subalpine fir trees in study sites between
1 to 2 years after severe drought. The high resistance (smaller growth reduction after severe
drought) of subalpine fir to drought observed, was consistent with findings by Maherali et al.
(2004) that showed gymnosperms (e.g., conifers) have a great capacity to resist a drought which
is due to xylem cells configuration. Furthermore, Clark et al. (2016) found that conifer species
dominant in western North America, were well adapted to drought. It has been also suggested that
under drier conditions, trees tend to increase belowground biomass allocation to improve their
water foraging capacity to maintain an efficient water transfer from soil to leaves (Brunner et al.
2015; Sala et al. 2012). This process requires substantial metabolic energy, since tree transpiration
is largely reduced during drought (Huang et al. 2018). As a result, drought forces trees to invest
large amounts of carbohydrates in metabolism, rather than in to radial growth (Brunner et al. 2015;
Huang et al. 2018).
Although Anderegg et al. (2015) suggested conifers need more time to recover their growth after
a water shortage than deciduous forests, we found that mountainous subalpine fir forests recovered
more quickly. It has been shown that trees have evolved many strategies to sustain drought stress.
For example, stomatal closure may limit water stress when conditions are not favorable, and
restricted shoot growth may result in trees having higher root-to-shoot ratios and greater capacity
to take up water relative to the shoots that must be supported (Brunner et al. 2015). In addition,

114

trees have extensive secondary growth, which allows them to increase the thickness of their cell
walls and results in increased resistance against cavitation in the vascular tissues (Jacobsen et al.,
2005). In severe drought condition when thresholds are reached, however, tree population
especially at high-latitude regions may be in danger of extinction. In turn, low-latitude tree
population may expand northward (Gazol et al. 2016) as a strategy to deal with drought. There are
many reasons why tree’s growth decreases in dry conditions. One of important reasons is lack of
nutrient supply due to reduced decomposition and mineralization, which are both dependant on
moisture and water supply (Prestch et al. 2013). There are variety of forest response strategies
against drought, and while some trees show greater growth resistance to drought, other show better
capacity to recover after drought (resilience). Forest response to drought sometimes depends on
site characteristics like soil texture and water table depth (Phillips et al. 2016). Some specific
ecophysiological responses may influence the way trees show resilience to severe climate
condition. However, according to Haung et al. (2018) geographical changes along with soil water
gradients play more important role for forest tree responses to drought. Huang et al. (2018) findings
support my results that subalpine fir showed different resilience to drought with changes in
latitude. Some site condition parameters which may change with latitude, and are important in
forest response to drought, include forest composition, soil type or depth, site size and age structure
site conditions (Gazol and Camarero 2016; Gazol et al. 2017b).
3.6. Conclusion
In this study, using tree-ring width data for the last six-seven decades, I presented evidence that
severe droughts caused growth reductions and legacies in our study stands. The magnitude of
legacy effects differed depending on the latitude. We observed that subalpine fir trees growing at
higher latitudes experienced lower disturbance impact (higher resistance) to drought than those
growing in lower latitude. Conversely, trees growing at lower latitude, returned faster to their pre115

disturbance state (higher recovery rate) than those growing at the higher latitudes, but the
mechanisms underlying these differences still remain unclear.
Overall, subalpine fir was resilient to drought at the sites investigated. Our analyses together with
previous findings indicate that different drought legacy effects on different subalpine fir stands
may be linked to the interactive roles of geographical changes, soil moisture gradients, water
availability and plant ecophysiological traits results in diverse lagged effects in response to severe
drought.
Finding of this study Chapter can be useful for drought risk assessment. Also, this finding could
guide forest practitioners in silvicultural practice by directly helping to choose silvicultural proper
methods like thinning and removing low resilient species to drought conditions in northern BC.
Furthermore, knowledge of subalpine fir recovery period and resilience can help forest
management through regulating sustainable long-term forest products like plantation of proper and
more drought-resilient species and also felling cycle regulations.
We suggest that more research and understanding of the interaction between drought legacy effects
and forest drought resilience is needed to better predict forest ecosystem responses to a probable
warmer and drier climate in the future.
Further experiments and studies on other elevations such as timberline and/or low-elevation forests
are also needed to test legacy effects and resilience of subalpine fir in western north Canada. Also,
further work should seek physiological and ecophysiological mechanisms involved in droughtinduced growth decline and how they can be influenced by specific species. This will ultimately
allow us to more confidently predict which species, in terms of their water use strategies, and
depending on site-specific climatic and edaphic characteristics, will be most beneficial for
increasing the resistance of forest stands to drought.

116

3.7. References:
Allen CD, Breshears DD, McDowell NG. 2015. On underestimation of global vulnerability to tree mortality
and forest die-off from hotter drought in the Anthropocene. Ecosphere, 6: 129.
Allen RG, Pereira LS, Raes D, Smith M. 1998. Crop evapotranspiration - Guidelines for computing crop
water requirements – FAO Irrigation and drainage paper 56. Food and Agriculture Organization of the
United Nations, Rome.
Allen CD, Macalady AK, Chenchouni H, Bachelet D, McDowell N, Vennetier M, Kitzberger T, Rigling A,
Breshears DD, Hogg EH, Gonzalez P, Fensham R, Zhang Z, Castro J, Demidova N, Lim JH, Allard G,
Running SW, Semerci A, Cobb N. 2010. A global overview of drought and heat-induced tree mortality
reveals emerging climate change risks for forests. Forest Ecology and Management. 259: 660-684
Anderegg LDL, and HilleRisLambers J. 2016. Drought stress limits the geographic ranges of two tree
species via different physiological mechanisms. Global Change Biology. 22: 1029–1045
Anderegg W R L, Schwalm C, Biondi F, Camarero JJ, Koch G, Litvak M, et al. 2015. Pervasive drought
legacies in forest ecosystems and their implications for carbon cycle models. Science. 349: 528-532 - doi:
10.1126/science.aab1833
Anderegg WRL, Plavcov AL, Anderegg LDL, Hacke UG, Berry JA, Field CB. 2013. Drought’s legacy:
multiyear hydraulic deterioration underlies widespread aspen forest die-off and portends increased future
risk. Global Change Biology. 19: 1188–1196.
Anderegg, WRL, Berry, JA, Smith DD, Sperry JS, Anderegg LD, Field CB. 2012. The roles of hydraulic
and carbon stress in a widespread climate-induced forest die-off. Proceedings of the National Academy of
Sciences. 109(1): 233-237.
Brodribb TJ, McAdam SAM, Jordan GJ, Martins SCV. 2014. Conifer species adapt to low-rainfall climates
by following one of two divergent pathways. Proceedings of the National Academy of Sciences of the USA.
111: 14489-14493.
Brunner I, Herzog C, Dawes MA, Arend M, Sperisen C. 2015. How tree roots respond to drought. Front.
Plant Science. 6: 547 - doi.org/10.3389/fpls.2015.00547
Clark JS, Iverson L, Woodall CW, Allen CD, Bell DM, Bragg DC, D’Amato AW, Davis FW, Hersh MH,
Ibanez I, Jackson ST, Matthews S, Pederson N, Peters M, Schwartz MW, Waring KM, Zimmermann NE.
2016. The impacts of increasing drought on forest dynamics, structure, and biodiversity in the United States.
Global Change Biology. 22: 2329-2352.
Carpenter S, Walker B, Anderies JM, Abel N. 2001. From Metaphor to Measurement: Resilience of What
to What? Ecosystems. Springer-Verlag :765-781 - doi.org/10.1007/s10021-001-0045-9
Camarero JJ, Gazol A, Sang€uesa-Barreda G, Oliva J, Vicente-Serrano SM. 2015 To die or not to die:
early-warning signals of dieback in response to a severe drought. Journal of Ecology. 103: 44-57.
Corcuera L, Camarero JJ, Gilpelegrin E. 2004. Effects of a severe drought on Quercus ilex radial growth
and xylem anatomy. Trees. 18: 83- 92.

117

Carrer M, and Urbinati C. 2004. Age-dependent tree-ring growth responses to climate of Larix decidua and
Pinus cembra in the Italian Alps. Ecology. 85: 730-740.
Cook ER, and Peters K. 1981. The smoothing spline: A new approach to standardizing forest interior treering width series for dendroclimatic studies. Tree-Ring Bulletin. 41: 45-53.
Cook ER, and Holmes RI. 1999. User’s manual for program ARSTAN. Tucson, AZ.
Cremonese E, Filippa G, Galvagno M, Siniscalco C, Oddi L, di Cella UM, et al. 2017. Heat wave hinders
green wave: the impact of climate severe on the phenology of a mountain grassland. Agricultural and Forest
Meteorology. 247: 320–330 - doi: 10.1016/j.agrformet.2017.08.016
Green RN.1994. Field guide for site identification and interpretation for the Vancouver forest region. Land
management handbook, No. 28.
DeLong C, Tanner D, Jull MJ, 1996. Draft field guide inserts for site identification and interpretation for
the southeast portion of the Prince George Forest Region. Ministry of Forests, Prince George, BC, Canada.
Delong C, Tanner D, Jull MJ. 1993. A field guide for site identification and interpretation for the southwest
portion of the Prince George forest region. Province of British Columbia: Ministry of Forests:
https://www.for.gov.bc.ca/hfd/pubs/docs/Lmh/Lmh24.pdf.
Dale MRT, and Fortin MJ. 2014. Spatial analysis: a guide for ecologists. Cambridge University Press.
De Boeck HJ, Bassin S, Verlinden M, Zeiter M, Hiltbrunner E. 2016. Simulated heat waves affected alpine
grassland only in combination with drought. New Phytologist. 209: 531-541 - doi: 10.1111/nph.13601.
Ettl GJ, and Peterson DL. 1995. Growth response of subalpine fir (Abies lasiocarpa) to climate in the
Olympic Mountains, Washington, USA. Global Change Biology. 1(3): 213-230.
Evans R, and Ilic J. 2001. Rapid prediction of wood stiffness from microfibril angle and density. Forest
Products Journal. 51(2): 53.
Gazol A, and Camarero JJ. 2016. Functional diversity enhances silver fir growth resilience to an severe
drought. Journal of Ecology. 104(4):1063-1075.
Gazol A, Camarero JJ, Anderegg WRL, Vicente-Serrano S. M. 2017a. Impacts of droughts on the growth
resilience of Northern Hemisphere forests. Global Ecology and Biogeography. 26: 166–176 - doi:
10.1111/geb.12526
Gazol A, Ribas M, Gutiérrez E, Camarero JJ. 2017b. Aleppo pine forests from across Spain show droughtinduced growth decline and partial recovery. Agricultural and Forest. Meteorology. 232: 186-194 - doi:
10.1016/j.agrformet.2016.08.014.
Gazol A, Sangüesa-Barreda G, Granda E, Camarero, JJ. 2017c. Tracking the impact of drought on
functionally different woody plants in a Mediterranean scrubland ecosystem. Plant Ecolology. 218: 1009–
1020 - doi: 10.1007/s11258-017-0749-3.
Gazol A, Camarero JJ, Gomez-Aparicio L. 2016. Functional diversity enhances silver fir growth resilience
to an severe drought. Journal Ecology. 104: 1063–1075 - doi.org/10.1111/1365-2745.12575

118

Gobiet A, Kotlarski S, Beniston M, Heinrich G, Rajczak J, Stoffel M. 2014. 21st century climate change in
the European Alps—a review. Science of the Total Environoment. 493: 1138–1151 - doi:
10.1016/j.scitotenv.2013.07.050
Gunderson L. 2000. Ecological resilience: in theory and application. Annual Review of Ecology, Evolution
and Systmeatics. 31: 425-439.
Gonzalez P, Neilson, RP, Lenihan JM, Drapek RJ. 2010. Global patterns in the vulnerability of ecosystems
to vegetation shifts due to climate change. Global Ecology and Biogeography. 19: 755–768.
Grissino-Mayer HD. 2001. Evaluating crossdating accuracy: a manual and tutorial for the computer
program COFECHA. Tree-Ring Research. 57(2): 205–221.
Greenwood S, Ruiz-Benito P, Martínez-Vilalta J, Lioret F, Kitzberger T, Allen CD, et al. 2017. Tree
mortality across biomes is promoted by drought intensity, lower wood density and higher specific leaf area.
Ecology Letters. 20: 539–553 - doi: 10.1111/ele.12748
Hacke U G, Stiller V, Sperry JS, Pittermann J, Mcculloh KA. 2001. Cavitation fatigue. Embolism and
refilling cycles can weaken the cavitation resistance of xylem. Plant Physiology. 125: 779–786.
Hinckley TM, Teskey RO, Waring RH, Morikawa Y. 1982. The water relations of true firs. In True Fir,
Proceedings of the Biology and Management of True Fir in the Pacific Northwest Symposium,
Contribution, Vol. 45: 85-92.
Jacobsen AL, Ewers FW, Pratt RB, Paddock WA, Davis SD. 2005. Do xylem fibres affect vessel cavitation
resistance? Plant Physiology. 139: 546-56 - doi.org/10.1104/pp.104.058404
Kienast F, Schweingruber FH, Bräker OU, Schär E. 1987. Tree-ring studies on conifers along ecological
gradients and the potential of single-year analyses. Canadian Journal of Forest Research. 17(7): 683-696.
Kannenberg SA, Maxwell JT, Pederson N, D'Orangeville L, Ficklin DL, Phillips RP. 2019. Drought
legacies are dependent on water table depth, wood anatomy and drought timing across the eastern
US. Ecology letters. 22(1): 119-127
Kuuluvainen T, Sprugel DG, Brooks J.R. 1996. Hydraulic architecture and structure of Abies lasiocarpa
seedlings in three subalpine meadows of different moisture status in the eastern Olympic mountains,
Washington, USA. Arctic and Alpine Research. 28(1): 60-64.
Lloret F, Keeling EG, Sala A. 2011. Components of tree resilience: effects of successive low-growth
episodes in old ponderosa pine forests. Oikos. 120: 1909-1920.
Lopez-Iglesias B, Villar R, Poorter L. 2014. Functional traits predict drought performance and distribution
of Mediterranean woody species. Acta Oecologica. 56: 10-18.
Lopez R, Cano FJ, Choat B, Cochard H, Gil L. 2016. Plasticity in vulnerability to cavitation of Pinus
canariensis occurs only at the driest end of an aridity gradient. Frontiers in Plant Science. 7: 769.
Maherali H, Pockman WT, Jackson RB. 2004. Adaptive variation in the vulnerability of woody plants to
xylem cavitation. Ecology. 85: 2184-2199 - doi: 10.1890/02-0538

119

Meko DM, and Baisan CH. 2001. Pilot study of latewood-width of conifers as an indicator of variability of
summer rainfall in the North American monsoon region. International Journal of Climatology. 21(6): 697708 - doi.org/10.1002/joc.646
Martin-Benito D, and Pederson N. 2015. Convergence in drought stress, but a divergence of climatic drivers
across a latitudinal gradient in a temperate broadleaf forest. Journal of Biogeography. 42: 925-937.
Millar CI, and Stephenson NL. 2015. Temperate forest health in an era of emerging mega disturbance.
Science. 349: 823-826.
Ogle K, Barber JJ, Barron-Gafford GA, Bentley LP, Young JM, Huxman TE, et al. 2015. Quantifying
ecological memory in plant and ecosystem processes. Ecology Letters. 18: 221-235 - doi:
10.1111/ele.12399
Pederson N, Dyer JM, McEwan RW, Hessl AE, Mock CJ, Orwig DA, Rieder DA, Cook BI. 2014. The
legacy of episodic climatic events in shaping temperate, broadleaf forests. Ecological Monographs. 599620, doi:10.1890/13-1025.1.
Peltier DMP, Fell M, Ogle K. 2016. Legacy effects of drought in the southwestern United States: a multispecies synthesis. Ecological. Monographs. 86: 312-326 - doi: 10.1002/ecm.1219
Pretzsch H, Schütze G, Uhl E. 2013. Resistance of European tree species to drought stress in mixed versus
pure forests: Evidence of stress release by inter-specific facilitation. Plant Biology. 15: 483-495 doi.org/10.1111/j.1438-8677. 00670.x
Peterson DW, Peterson DL, Ettl GJ. 2002. Growth responses of subalpine fir to climatic variability in the
Pacific Northwest. Canadian Journal of Forest Research. 32(9): 1503-1517.
Peng C, Ma Z, Lei X, Zhu Q, Chen H, Wang W, Liu S, Li W, Fang X, Zhou X. 2011. A drought-induced
pervasive increase in tree mortality across Canada’s boreal forests. Nature Climate Chang. 1: 467-471 doi.org/10.1038/nclimate1293
Peltier DMP, Fell M, Ogle K. 2016. Legacy effects of drought in the southwestern United States: a multispecies synthesis. Ecological. Monographs. 86: 312–326 - doi: 10.1002/ecm.1219
Phillips RP, Ib_a~nez I, D’Orangeville L, Hanson PJ, Ryan MG, McDowell NG. 2016. A belowground
perspective on the drought sensitivity of forests: towards improved understanding and simulation. Forest
Ecology and Management. 380: 309-320.
Pourtahmasi K, Lotfiomran N, Br€auning A, Parsapajouh D. 2011. Tree-ring width and vessel
characteristics of oriental beech (Fagus orientalis L.) along an altitudinal gradient in the Caspian forests,
northern Iran. IAWA Journal. 32: 461 473.
Qin D, Plattner GK, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM. 2014. Climate
change 2013: the physical science basis. Contribution of Working Group I to the Fifth Assessment Report
of the Intergovernmental Panel on Climate Change. 5-14.
Reyer CPO, Brouwers N, Rammig A, Brook BW, Epila J, Grant RF, Holmgren M, Langerwisch F,
Leuzinger, S, Lucht W, Medlyn B, Pfeifer M, Steinkamp J, Vanderwel MC, Verbeeck H, Villela DM. 2015.
Forest resilience and tipping points at different spatio-temporal scales: Approaches and challenges. Journal
of Ecology. 103: 5-15 - doi.org/10.1111/1365-2745.12337
120

Reyer C, Lasch-Born P, Suckow F, Gutsch M, Murawski A, Pilz T. 2014. Projections of regional changes
in forest net primary productivity for different tree species in Europe driven by climate change and carbon
dioxide. Annals of Forest Science. 71: 211-225- doi. org/10.1007/s1359 5-013-0306-8
Rozas V. 2005. Dendrochronology of pedunculate oak (Quercus robur L.) in an old-growth pollarded
woodland in northern Spain: tree-ring growth responses to climate. Annals of Forest Science. 62: 209-218.
Rossi S, Deslauriers A, Anfodillo T, Carrer M. 2008. Age-dependent xylogenesis in timberline conifers.
New Phytologist. 177: 199-208.
Sala A, Woodruff DR, Meinzer FC. 2012. Carbon dynamics in trees: feast or famine? Tree Physiology. 32:
764-775 - doi.org/10.1093/treephys/tpr143
Sánchez-Salguero R, Navarro-Cerillo RM, Camarero JJ, Fernández-Cancio A. 2012. Selective droughtinduced decline of pine species in southeastern Spain. Climate Change. 113: 767-785 - doi:
10.1007/s10584-011-0372-6
Seidl R, Rammer W, Spies TA. 2014. Disturbance legacies increase the resilience of forest ecosystem
structure, composition, and functioning. Ecological. Applications. 24: 2063- 2077 - doi.org/10.1890/140255.1
Schwalm CR, Anderegg WRL, Michalak AM, Fisher JB, Biondi F, Koch G, Litvak M, Ogle K, Shaw JD,
Wolf A, Huntzinger DN, Schaefer K, Cook R, Wei Y, Fang Y, Hayes D, Huang M, Jain A, Tian H. 2017.
Global patterns of drought recovery. Nature. 548: 202–205 - doi.org/10.1038/nature23021
Steinkamp J, and Hickler T. 2015. Is drought-induced forest dieback globally increasing? Journal of
Ecology. 103: 31-43.
Stephenson NL. 1998. Actual evapotranspiration and deficit: biologically meaningful correlates of
vegetation distribution across spatial scales. Journal of Biogeography. 25: 855-870.
Vicente-Serrano SM, Gouveia C, Camarero JJ, Beguería S, Trigo R, López-Morenoa JI, et al. 2013.
Response of vegetation to drought timescales across global land biomes. Proceeding of National Academy
of the Sciences. U.S.A. 110: 52–57 - doi: 10.1073/pnas.1207068110
Virlouvet L and Fromm M. 2015. Physiological and transcriptional memory in guard cells during repetitive
dehydration stress. New Phytologist. 205: 596–607- doi.org/10.1111/nph.13080
Wigley TML, Briffa KR, Jones PD. 1984. On the average value of correlated time series, with applications
in dendroclimatology from hotter drought in the anthropocene. Ecosphere. 6: 1-55.
Wu X, Liu H, Li X, Ciais P, Babst F, Guo W, et al. 2018. Differentiating drought legacy effects on
vegetation growth over the temperate Northern Hemisphere. Global Change Biology. 24: 504-516 - doi:
10.1111/gcb.13920
Yamaguchi DK. 1991. A simple method for cross-dating increment cores form living trees. Canadian
Journal of Forest Research. 21: 414-416.
Zweifel R, and Sterck F. 2018. A conceptual tree model explaining legacy effects on stem growth. Frontiers
in Forests and Global Change. 1: 9.

121

Zhang Q, Shao M, Jia X, Wei X. 2017. Relationship of climatic and forest factors to drought and heatinduced tree mortality. PLoS ONE. 12:ve0169770 - doi: 10.1371/journal.pone.0169770
Zhang Q, Alfaro RI, Hebda RJ. 1999. Dendroecological studies of tree growth, climate and spruce beetle
outbreaks in Central British Columbia, Canada. Forest Ecology and Management. 121(3): 215-225.

122

Chapter 4: Concluding Synthesis
This thesis aimed to determine how subalpine fir (Abies lasiocarpa) wood properties including
RW, EW, LW, RD, FL, MFA, FW and FWT vary over time under the impacts of climate in the
Rocky Mountains in northern British Columbia. The reason why I chose mountainous high
elevation forests was because it is shown that global warming significantly affects ecosystems of
high latitudes and elevations as mountainous high-elevated forests are more sensitive to climate
variations (Becker and Bugmann 2001; Frank and Esper 2005; Anderson and Goulden 2011;
Tognetti and Palombo 2013). This thesis also aimed to determine growth resilience (recovery
period) and legacy effects in subalpine fir stands after drought conditions, and determining how
wood properties and resilience of subalpine fir vary among the study sites and along various
latitudes in dealing with climate change.
4.1 KEY POINTS
What can be concluded from research in Chapter 2 is that subalpine fir’s radial growth at study
sites is more likely to depend on variations in temperature and less so on precipitation, on northern
aspects of the Rocky Mountains in northern BC. Moreover, our results illustrated that, compared
to simple tree-ring width, intra-annual-ring characteristics generally showed stronger and more
robust correlations with climate variations as have been studied by other researchers (Wimmer and
Grabner 2000; Davi et al. 2002). This is likely because tree-ring estimations give just a growing
season portrayal of climate information (D’Arrigo et al. 1993). However, to measure and
understand exact climatic and ecological changes over smaller periods of time, intra-annual wood
property estimations provide stronger relationships to climate than do RW values (Wimmer and
Grabner 2000; Davi et al. 2002).

123

In this thesis and specifically in Chapter 2, I used climate and wood chronology data from over
80 years to study subalpine fir’s fibre property responses to climate variations (temperature and
precipitation). The period from 1940 to 2018 showed that summer temperature, and to a lesser
extent previous fall precipitation, controlled fibre properties growth at higher latitudes. I also
concluded that the lower the latitude, the greater the effect of fall temperature on fibre properties
at corresponding sites. These finding are logical as water is typically not a limiting factor at higher
latitudes due to colder weather and larger snowpack. Furthermore, inter / intra-annual-ring
properties appeared to be more influenced by spring and summer temperature and to a lesser extent
by spring and late summer precipitation at study sites in northern BC.
Climatic variants including mean temperature and sum of annual precipitation have been increased
over four decades from 1940s to 1970s, which resulted in increased fibre length in study sites.
Following that and during 1980s and 1990s, higher than average temperature and lower
precipitation, which to the larger scale were influenced by macro-climatic phenomena like springsummer PDO thermal phases, decreased subalpine fir fibre length from elongation at study sites.
However, in 2000s and 2010s, increases in the amount of precipitation along with increased
temperatures were probably the reasons for producing larger fibre lengths at study sites in northern
BC and along northern aspects of the Rockies. I observed increases in fibre length at sites located
at lower latitudes compared with ones at higher latitude during 1940 to 2018, possibly due to longer
growing period compared to the higher latitude stands where growing periods were shorter.
In Chapter two and studying wood properties, we also came to the conclusion that maximum
density was positively affected by late summer precipitation at all study sites other than at Averil,
over recent 80 years. Because the study sites were located at high altitudes (1100 m-1300 m a.s.l),
probably the rise in temperature late in summer over last 7-8 decades along with sufficient

124

humidity in study sites proved optimal conditions for formation of latewood cells leading to denser
latewood cells. A possible reason why subalpine fir MXD did not increase with summer
precipitation at the Averil site could be because summer precipitation exceeded soil water
availability at this site; Averil was wetter than all other study sites, receiving higher amount of
precipitation during the year.
In Chapter 3, I found that drought events during 1940 to 2018, reduced subalpine fir radial growth
in our study sites along northern slopes of the Rocky Mountains. Following that, I focused on the
periods between 1950 and 2018, during which we observed few severe dry years, with 1 to 2 lags
in growth after drought which obviously were important to restore ecosystem health after dry
periods.
On one hand, in chapter 3 we found that at higher latitudes subalpine fir was more resistant to
severe drought than the ones at lower latitudes. On the other hand, subalpine fir growing at lower
latitudes showed higher recovery rate (more resilience) than those growing at higher latitudes.
Also, we concluded that at higher latitudes, legacy effects took 2 years to compensate declined
growth which likely suggest that resilience of subalpine fir trees at higher elevations was lower
than at lower latitudes where legacy effects took only 1 year to return to pre-perturbation
conditions after severe drought (recovery period). Increase in resistance to drought in forest stands
at higher latitudes were also studied by Gazol et al. (2016) which might be due to them receiving
more precipitation as well as better soil moisture regimes at higher latitudes and altitudes. Our
results for fast drought recovery at lower latitudes, can be supported by Gazol et al. (2016) who
confirmed that higher values of forest growth recovery existed in the western south USA and
southern Europe at lower latitudes.

125

Although the exact physiological mechanisms associated with the response of trees to severe
drought remain unclear, Gazol et al. (2016) suggested the existence of different strategies in forest
trees to cope with droughts, including: change in the biome, geography, genetic adaptation, soil
moisture gradient, water availability, soil texture and forest composition. Temperature changes at
different latitudes may result in change in the sensitivity of trees to drought as Briffa et al. (1998)
suggested a reduction in sensitivity of trees growth at high northern latitudes for several decades.
Overall, all things considered in Chapter 2 and 3, and according to the correlations and regressions
analysis, I came to the general conclusion that temperature was the most important climatic factor
affecting growth and resilience of subalpine fir in the study sites. It worth mentioning than although
drought is consequence of interaction (balance) between lower precipitation and higher
temperature, still at higher elevations of the Rocky Mountains in Northern BC temperature controls
the growth of subalpine fir. Also, subalpine fir was resilient to drought at the sites investigated
along northern slopes of the Rocky Mountains in northern BC.
ClimateBC note
The most reliable way to study the behavior of trees in relation to climate is to use long-term fixed
(historical) meteorological data. However, when long-term climate data is not available or
meteorological stations are not available at or close to the sampling site, the best way to collect
climate data is to use normal and standard reliable meteorological models. There are some pros
and cons for using models, and this research was done knowing the limitations of one such model
in BC, ClimateBC. ClimateBC uses information of historical climate station in BC to match the
best model (Wang et al. 2016). ClimateBC model producers used ClimateWNA (version 4.6)
(Wang et al. 2012) to generate climate data. ClimateWNA also downscales historical and future
climate data, and outputs monthly, seasonal and annual temperature and precipitation variables, as

126

well as derived annual climate variables of biological significance to plants (Wang et al. 2012).
ClimateBC model is made using 20 reliable climate changes models (Wang et al. 2012). The
program uses the scale-free data as baseline in combination with monthly anomaly data (Mitchell
and Jones 2005) of individual years to calculate historical monthly, seasonal and annual climate
variables for individual years and periods between 1901 and 2012 (Wang et al. 2012).
However, model accuracy remains major challenges for incorporating these projections into
climate change adaptation strategies (Dormann et al. 2008; Wang et al. 2012). Furthermore,
climate models of ecosystem change have been criticized for alterations to biogeochemical cycles,
including increased atmospheric CO2 concentrations (Pearson and Dawson 2003; Araujo and
Guisan 2006).
Accuracy of ClimateBC has been tested by model makers. To test whether the site-specific climate
data generated by ClimateBC was likely to share the same relationship derived from another
researcher report at Agassiz station, Wang et al. (2012) compared modelled monthly temperature
and precipitation climate records to the Agassiz station data. Correlation analyses showed that
mean summer and winter temperature values from the two data sets were significantly correlated
(r = 0.841, p < 0.01; and r = 0.905, p < 0.01, respectively).
Considering all pros and cons for using this model, ClimateBC still is one the best models for
predicting climate variations in BC and has been applied in many studies so far (Canno et al. 2012;
Brochett et al. 2012; Spittlehouse and Wang. 2014; Pidwirny et al. 2019; Bahbahani and Pidwirny.
2017; Cortini et al. 2011; Meyn et al. 2010; Rose and Burton 2009; Delong et al. 2009).

4.2. CONSIDERATIONS FOR FUTURE RESEARCH

127

This research is a preliminary study of relationship between climate and wood anatomical
properties and residence of subalpine fir located in the Rocky Mountains of northern interior
British Columbia, Canada. Future research into relationships between climate and radial growth
properties (RW, EW, LW, RD), wood fibre properties (FL, FW, FWT, MFA), and drought legacy
effect studies could investigate additional climate variables, such as snow depth, relative humidity,
wind speed and number of cloudy days in the year to improve models of radial growth and fibre
properties based on climate. Investigation into variables in addition to climate, such as crown
cover, forest composition, forest species could also improve estimations of variations in radial
growth and resilience of forest trees. Also, further work is needed to clarify the physiological
mechanisms involved in drought-induced growth decline and how they can be influenced by intra/ inter-species interactions and forest management.
Further efforts identifying which combinations of species mixtures, in terms of their water use
strategies, and depending on site specific climatic and edaphic characteristics, will be most
beneficial for increasing the resilience of forests to drought.
Finally, an indispensable requirement for assessing the long-term impact of drought events on trees
is the availability of long-term records (Bräuning et al. 2017). Therefore, by having several periods
of drought over a long period of time, we can measure a better average for the degree of flexibility
of trees in response to perturbations. Also, ensuring adequate number of samples (sections) for
wood fibre analysis vs 10-year average sections allow us for better and more precise fibre
measurement which is lacking in our research. Furthermore, lack of significant relationships
between climate indices (eg., drought) and RW, EWW at Grizzly Den stands was likely due to
small sample size in this study. Due to time constraints and research costs, this study was
conducted in five regions with different latitudes. It is suggested that more study sites be

128

considered in different latitudes and altitudes gradient for future research as well as more study
sites in each latitude which were lacking in this research. Also, in this study, no statistical
comparison was made based on the significant correlation between the studied sites. We suggest
to compare more statistical correlations between chronologies and climate between sites in future
studies.
4.3. IMPLICATIONS FOR THE BRITISH COLUMBIA FORESTRY INDUSTRY
Examination of the general relationships between climate and radial growth and wood fibre
variables may answer the questions regarding wood quality in subalpine fir stands of Prince
George. It was demonstrated that over the recent 70 to 80 years, rising temperatures and variation
in annual precipitation in Prince George were negatively, correlated with radial growth (RW) and
slightly related to increased ring density and production of cells that were thicker and smaller in
study stands. Producing denser wood with smaller annual-ring width may be ideal for industrial
solid wood production. However, decreasing trends in precipitation, and rises in average
temperature generally in study stands and over recent 70 to 80 years have reduced production of
high-volume subalpine fir’s wood for high-volume logs and end-use industrial purposes.
Subalpine fir trees in mountainous study stands in northern Rockies showed a relatively short
recovery periods from drought conditions meaning that they are fairly resilient to drought (1-2
years legacy effects) and are well adapted (especially at McBride Peak site) to drought with respect
to other environmental factors and regional climate which probably help to produce sustainable
forest products future for BC and provide data for improving the Government of BC Stand Level
Drought Risk Assessment Tool. Also, increased awareness of subalpine fir fibre property changes
under global climate changes and regarding that fibre length have been increased in study sites
over recent 70 to 80 years, could feed directly into knowledge of supply quality for end-uses such
as pulp and paper in BC.
129

4.4. References
Araujo MB, Guisan A. 2006. Five (or so) challenges for species distribution modelling. Journal of
Biogeography. 33: 1677-1688.
Anderson RG, and Goulden ML. 2011. Relationships between climate, vegetation, and energy exchange
across a mountain gradient. Journal of Geophysical Research: Biogeosciences. 116(G1) doi.org/10.1029/2010JG001476
Arend M, and Fromm J. 2007. Seasonal change in the drought response of wood cell development in poplar.
Tree Physiology. 27: 985-992 - doi: 10.1093/treephys/27.7.985
Badel E, Ewers FW, Cochard H, Telewski FW. 2015. Acclimation of mechanical and hydraulic functions
in trees: impact of the thigmomorphogenetic process. Frontiers in Plant Science 6 - doi:
10.3389/fpls.2015.00266
Bahbahani KM, and Pidwirny MJ. 2017. December. Spatial Analysis of the Effects of the Anomalous
Winter of 2014/15 on 157 Ski Resorts Located in Western North America. In AGU Fall Meeting
Abstracts (Vol. 2017: A53C-2259).
Barnett JR, and Jeronimidis (eds). 2003. Tree growth and wood quality. Blackwell Publishing: Oxford
(UK).
Becker A, and Bugmann H (eds). 2001. Global Change and Mountain Regions: The Mountain Research
Initiative. IGBP Report 49, Stockholm. p 48.
Brockett BF, Prescott CE, Grayston SJ. 2012. Soil moisture is the major factor influencing microbial
community structure and enzyme activities across seven biogeoclimatic zones in western Canada. Soil
biology and biochemistry. 44(1): 9-20.
Briffa, KR, Schweingruber FH, Jones PD, Osborn TJ, Shiyatov SG and Vaganov EA. 1998. Reduced
sensitivity of recent tree-growth to temperature at high northern latitudes. Nature. 391(6668): 678-682.
Bräuning A, Bolte A, Nabais C, Rossi S, Sass-Klaassen U. 2017. Editorial: Studying Tree Responses to
Severe Events. Front. Plant Science. 8, 506 - doi.org/10.3389/fpls.2017.00506
Cannon AJ, Neilsen D, Taylor B. 2012. Lapse rate adjustments of gridded surface temperature normal in
an area of complex terrain: atmospheric reanalysis versus statistical up-sampling. Atmosphereocean, 50(1): 9-16.
Cortini F, McClarnon J, Bedford L, Comeau PG, Powelson A, and Boateng JO. 2011. Can site preparation
help lodgepole pine and white spruce deal with climate change?
Davi NK, D’Arrigo RD, Jacoby JG, Buckley B, Kobayashi O. 2002. Warm-season annual to decadal
temperature variability for Hokkaido, Japan, inferred from maximum latewood density (AD 1557–1990)
and ring width (AD 1532–1990). Climate Change. 52:210-217.
Delong C. Griesbauer H, Mackenzie W, and Foord V. 2009. Corroboration of biogeoclimatic ecosystem
classification climate zonation by spatially modelled climate data. Journal of Ecosystems and
Management. 10(3).

130

Dormann CF, Purschke O, Marquez JRG, Lautenbach S, Schroder B. 2008. Components of uncertainty in
species distribution analysis: a case study of the great grey shrike. Ecology. 89: 3371-3386.
Frank D, and Esper J. 2005. Characterization and climate response patterns of a high-elevation, multispecies tree-ring network in the European Alps. Dendrochronologia. 22:107–121 - doi:
10.1016/j.dendro.2005.02.004
Gazol A, Camarero JJ, Gomez-Aparicio L. 2016. Functional diversity enhances silver fir growth resilience
to a severe drought. Journal of Ecology. 104: 1063–1075 - doi.org/10.1111/1365-2745.12575
Hacke UG, and Sperry JS. 2001. Functional and ecological xylem anatomy. Perspectives in Plant Ecology,
Evolution and Systematics. 4:97-115.
Meyn A, Schmidtlein S, Taylor SW, Girardin MP, Thonicke K, and Cramer W. 2010. Spatial variation of
trends in wildfire and summer drought in British Columbia, Canada, 1920–2000. International Journal of
Wildland Fire. 19(3): 272-283.
Rossi S, Morin H, Deslauriers A, Plourde P-Y. 2011. Predicting xylem phenology in black spruce under
climate warming. Global Change Biology. 17: 614–625.
Rose NA, and Burton PJ. 2009. Using bioclimatic envelopes to identify temporal corridors in support of
conservation planning in a changing climate. Forest Ecology and Management, 258: S64-S74.
Schwalm CR, Anderegg WRL, Michalak AM, Fisher JB, Biondi F, Koch G et al. 2017. Global patterns of
drought recovery. Nature. 548: 202-205.
Spittlehouse D, and Wang T. 2014. Evaluation of ClimateBC V5.
Sperry JS, Hacke UG, Pittermann J. 2006. Size and function in conifer tracheids and angiosperm vessels.
American Journal of Botany. 93: 1490 -1500.
Tranquillini W. 1979. Physiological Ecology of the Alpine Timberline. Springer. Berlin- Heidelberg.
Tognetti R, and Palombo C. 2013. Take a tree to the limit: the stress line. Tree Physiology. 33(9):887-890
- doi.org/10.1093/treephys/tpt087
Pearson RG, Dawson TP. 2003. Predicting the impacts of climate change on the distribution of species: are
bioclimate envelope models useful? Global Ecology and Biogeography. 12: 361-371.
Peltier, DMP, Fell M, Ogle K. 2016. Legacy effects of drought in the southwestern United States: a multispecies synthesis. Ecological Monographs. 86: 312–326 - doi: 10.1002/ecm.1219
Pidwirny MJ, Clark E, Bahbahani KM. 2019. Climate Change Challenges for Alpine Ski Resorts in Western
Canada: Future Predicted Changes in Winter Mean Temperature, Winter Snowfall, and Ski Season Length.
In AGU Fall Meeting Abstracts (Vol. 2019: GC44B-06).
Von Arx G, Archer SR, Hughes MK. 2012. Long-term functional plasticity in plant hydraulic architecture
in response to supplemental moisture. Annals of Botany. 109: 1091-1100 - doi: 10.1093/aob/mcs030

131

Wang T, Hamann A, Spittlehouse D, Carroll C. 2016. “Locally Downscaled and Spatially Customizable
Climate Data for Historical and Future Periods for North America.” PLoS ONE. 11(6): e0156720.
Wang T, Hamann A, Spittlehouse D, Murdock TN. 2012. ClimateWNA – High-resolution spatial climate
data for western North America. Journal of Applied Meteorology and Climatology. 61: 16-29.
Wimmer R, and Grabner M. 2000. A comparison of tree-ring features in Picea abies as correlated with
climate. IAWA Journal. 21: 403- 416.

132

Appendix 1

Following graphs were made based on climate information extracted from ClimateBC by putting
the exact altitude and latitude information for each site (Wang et al. 2012; Spittlehouse and Wang
2014). To obtain stronger models, multi variable model can be applied using two or more
climatic variables and putting in the same model to predict better and stronger models provided
that the independent variables (climatic parameters) are not statistically correlated. To find if
variables were related together or not, a Durbin-Watson statistic was applied. If independent
variables were autocorrelated we could potentially take an average of two independent variables
and put that average in the model for prediction. For those independent climate variables that
were different (e.g., temperature and precipitation), and needed to be combined in a model,
fluctuations over time were compared visually to make sure they have same trends.

Figure A1. Comparison between standardized precipitation of May, June, July, and temperature of September,
October, November and December at Averil site from 1940- 2018 along the northern slopes of the Rocky Mountains,
northern BC.

133

Figure A2. Comparison between standardized precipitation of May, September, October, November and December,
and temperature of May and June at Averil site from 1940- 2018 along the northern slopes of the Rocky Mountains,
northern BC.

Figure A3. Comparison between standardized August temperature and September precipitation trends at Pine Le
Moray site from 1950- 2018 along the northern slopes of the Rocky Mountains, northern BC.

Figure A4. Comparison between standardized previous fall precipitation and previous fall minimum temperature
trends at McBride Peak site from 1950- 2018 along the northern slopes of the Rocky Mountains, northern BC.

134

Figure A5. Precipitation as snow (PAS) at Pine Le Moray, Chingee, Averil, Grizzly Den and McBride Peak sites from
1950- 2018. Average snow depth at Pine Le Moray = 625mm, Chingee = 609 mm, Averil = 607mm, Grizzly Den =
561mm and at McBride Peak = 399mm along the northern slopes of the Rocky Mountains. Northern BC.

Figure A6. Number of days that temperature was higher than 18 ºC at Grizzly Den site from 1950-2018 along the
northern slopes of the Rocky Mountains, northern BC. Dashed trend line represents entire trend in the study site.

Figure A7. Previous July and August air temperature at McBride Peak from 1940 -2018. Dashed trend lines
represent entire climate history trend in the study site.

135

Table A1. Durbin-Watson statistic for climate variations at Pine Le Moray, Averil and McBride Peak sites along the
northern slopes of the Rocky Mountains, northern BC. Values ranging from 1.5 to 2.2, with of value of 2.0 representing
no autocorrelation. Values above 2.2 and less than 11.5 were deemed positively autocorrelated.

Site

Pine Le Moray

Wood characteristic

Durbin Watson
value

Average and Maximum temperature of May

Fibre wall thickness

0.88

Average temperature of August + Minimum temperature
of Jully and August

Micro fibril angle

2.21

Average temperature of June and Jully + Min temperature
of Jully

Fibre width

1.4

Minimum density

2.3

Fall and summer maximum temperature

Fibre length

1.14

precipitation of May, June, July and temperature of
September, October, November, December

Ring density

0.9

Precipitation of May, September, October, November,
December, and temperature of May, June

Latewood density

0.9

Minimum temperature and precipitation of previous fall

Fibre length

2.88

Earlywood width

2.43

Climate parameters

August temperature + September precipitation

Averil

McBride Peak

Temperature of previous Jully and August

136

References:
Spittlehouse D, and Wang T. 2014. Evaluation of ClimateBC V5.
Wang T, Hamann A, Spittlehouse DL, Murdock TQ. 2012. ClimateWNA-high-resolution spatial climate data for
western North America. Journal of Applied Meteorology and Climatology. 51(1): 16-29.

137