Access to this full-text is provided by Canadian Science Publishing.
Content available from Environmental Reviews
This content is subject to copyright. Terms and conditions apply.
OPEN ACCESS |Review
Future impacts of climate change on black spruce growth
and mortality: review and challenges
Jonathan A. Lesven a,b, Milva Druguet Dayrasa,c, Jonathan Cazabonnec, François Gilleta, André Arsenaultd,
Damien Riusa, and Yves Bergeronb,e
aLaboratoire Chrono-Environnement, UMR 6249 CNRS, Université de Franche-Comté, 16 Route de Gray, Besançon 25000, France;
bInstitut de Recherche sur les Forêts, Université du Québec en Abitibi-Témiscamingue, 445 Boulevard de l’Université,
Rouyn-Noranda, QC J9X 5E4, Canada; cGroupe de recherche en Écologie de la MRC-Abitibi (GREMA), Institut de Recherche sur les
Forêts, Université du Québec en Abitibi-Témiscamingue, 341 rue Principale Nord, Amos, QC J9T 2L8, Canada; dNatural Resources
Canada, Canadian Forest Service – Atlantic Forestry Centre, Corner Brook Oce, PO Box 960, 20 University Drive, Corner Brook, NL
A2H 6J3, Canada; eDépartement des sciences biologiques, Université du Québec à Montréal, CP 8888, Succursale Centre-Ville,
Montréal, QC H3C 3P8, Canada
Corresponding author: Jonathan A. Lesven (email: jonathan.lesven@uqat.ca)
Abstract
Black spruce (Picea mariana (Mill.) B.S.P.) is the dominant conifer species across a large part of North American boreal forests,
providing many goods and services essential to human activities, and playing a major climatic role through the global carbon
cycle. However, a comprehensive synthesis of the eects of climate change on black spruce has not yet been undertaken. The
dynamics of black spruce are influenced by various living (biotic) and non-living (abiotic) factors, as well as their combined
eects, which are particularly responsive to changes in climate. Climate change predictions suggest that northern ecosystems
will experience the world’s most significant impact. Therefore, black spruce is likely to undergo profound disruptions in its
growth and mortality rate in the next few decades, resulting in significant changes in forestry and carbon storage. However,
these changes will not be uniform throughout the entire distribution of the species. Future changes in temperature and
precipitation will create more stress for water availability in the boreal forests of western and central North America than
in their eastern counterparts. Thus, significant longitudinal and latitudinal gradients in tree growth and mortality variability
are expected throughout the range of the species. This literature review aims to summarise the impacts of climate change on
individual tree growth and mortality of this major species. While enhanced black spruce productivity could occur through
both increased air temperature and nitrogen mineralisation in the soil, moisture limitation in central and western North
America will result in significant growth reduction and mortality events across these regions. Conversely, under the expected
climate change scenarios, black spruce forests may be more resilient in eastern North America, where climatic conditions
appear more suitable, particularly in their northernmost range. In this review, we identify current research gaps for some
disturbances, which should be addressed to better understand the impact of climate change on black spruce. Finally, we
identify issues associated with sustainable forest management and the maintenance of black spruce under projected future
climate changes.
Key words: black spruce, growth, mortality, climate change, sustainable forest management
Introduction
Global temperature has increased by an average of 1 ◦C
since the preindustrial era, a pattern that is likely to be exac-
erbated as a result of human-induced rising atmospheric CO2
concentrations (hereafter [CO2]) (Field et al. 2014;Pörtner et
al. 2022). As the magnitude of future greenhouse gas emis-
sions is still uncertain, several scenarios have been devel-
oped (van Vuuren et al. 2011). For example, the RCP2.6 sce-
nario model predicts a +1.8 ◦C temperature increase at the
global scale under the lowest estimated 2100 [CO2] equiva-
lent, whereas the RCP8.5 scenario projects the highest tem-
perature increase, estimated at +4.3 ◦C, under the highest
estimated concentrations (Pörtner et al. 2022). Exponential
anthropogenic forcing is projected to have a tremendous im-
pact on northern biomes, estimated at more than twice the
global average (Field et al. 2014). This could reach up to 10 ◦C
compared to the pre-industrial era at high latitudes under
the RPC8.5 scenario, with precipitation also being expected
to increase by 20%–30% (Field et al. 2014;Pörtner et al. 2022).
The estimated temperature increases will result in higher
rates of evapotranspiration that will not be fully compen-
sated for, despite concurrent increases in precipitation (Tam
et al. 2018), leading to major vegetation shifts and mortal-
ity episodes. Shrinking individual growth, reduced stand den-
214 Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075
Canadian Science Publishing
Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075 215
sity, and the resulting large-scale boreal forest dieback have
consequently been defined as a tipping point, i.e., a critical
threshold at which a small perturbation in forcing can fun-
damentally and irreversibly alter the ecosystem state (Rao et
al. 2023). This threshold may be crossed under a global warm-
ing scenario of 3 ◦C(Lenton et al. 2008). In turn, global warm-
ing has the potential to considerably impact individual tree
growth and survival, resulting in either positive or negative
eects for both carbon storage (Kurz et al. 2008) and timber
supply (Gauthier et al. 2015). Our ability to anticipate the im-
pact of these changes is of major interest to land managers
and stakeholders in developing realistic targets and goals for
sustainable forest management strategies that promote the
long-term resilience of these ecosystems.
Black spruce (Picea mariana (Mill.) B.S.P.——hereafter PIMA)
is a conifer species endemic to North America (hereafter
NA) that is currently widely distributed throughout Canada,
the northeastern USA, and Alaska (Viereck et al. 1990). It
represents a major economic resource, especially in east-
ern and central Canada, as it is widely used for both wood
and paper pulp supply (Natural Resources Canada 2022b).
PIMA forests also serve multiple important societal func-
tions by being frequently utilised for spiritual, educational,
and recreational activities (Hassan et al. 2005). Additionally,
these forests hold significant traditional value for Indigenous
Peoples (Clément 1995) and have a crucial climatic role in
storing substantial amounts of biogenic carbon (Pan et al.
2011).
As a shallow-rooted species, PIMA is particularly sensitive
to water stress and grows mainly on nutrient-poor soils in
cold and humid to dry subhumid environments (Girardin et
al. 2016b), although it can also be found on more mesic sites.
As a fire-adapted species due to its semi-serotinous cones,
PIMA life-history at the landscape scale is also largely re-
lated to the historical occurrence and severity of fires, which
allows for its post-fire renewal and maintenance over time
(Johnstone et al. 2004). However, PIMA also reproduces by
layering, particularly in regions of high moisture where fires
are less frequent or where temperatures do not allow for vi-
able seed production, thus emphasising its wide ecological
range (Viereck et al. 1990). Future climate changes, through
their capacity to modify evaporative balances and distur-
bance regimes, have a significant potential to alter PIMA
growth and mortality rates, as well as its spatial distribution
and interactions with other species. These changes will ul-
timately result in a balance of eects that may or may not
be favourable to its maintenance in the future. For exam-
ple, longer growing seasons and warmer temperatures may
favour PIMA productivity and increase soil nutrient avail-
ability (Strömgren and Linder 2002;Bronson et al. 2009),
while its mortality rate may be reduced at the northern limit
of its distribution through the northward displacement of
isotherms (Gamache and Payette 2004). On the other hand,
enhanced water stress may limit its photosynthetic capacity
and consequently decrease its growth potential (Goetz et al.
2005). In addition, the mortality rate may increase by shifting
the range of insect pests and pathogen outbreaks to regions
where PIMA was previously protected by harsh winter condi-
tions (Régnière et al. 2012).
While these changes in the growth and mortality rate of
PIMA will have wide economic, ecological, and climatic im-
plications across North American ecosystems, they are pre-
dicted to be far from uniform, leading to significant regional
discrepancies in the severity and direction of the expected
impacts. Western and central NA are already experiencing
longer and more frequent droughts than eastern Canadian
regions (Bonsal et al. 2011), a pattern that is likely to be par-
ticularly exacerbated under high anthropogenic forcing sce-
narios (Tam et al. 2018). This increase will likely lead to en-
hanced wildfire activity in western and central NA (Flannigan
et al. 2005;de Groot et al. 2013), resulting in shifts from
PIMA-dominated to broadleaf-dominated forests, induced by
too short fire return intervals (Baltzer et al. 2021). Superim-
posed on these disturbance-mediated changes in forest stand
composition, permafrost thaw is also expected to lead to ge-
ographic changes in tree growth conditions that will vary ac-
cording to longitude, latitude, elevation, and expected distur-
bances (Duveneck et al. 2014).
Understanding the spatial variability of individual re-
sponses of PIMA trees to climate change is thus of major in-
terest for projections of future global carbon sequestration
and timber production, and to inform climate change mitiga-
tion and sustainable forest management strategies in north-
ern ecosystems. In this context, our review summarises the
main findings from the last decades on the major abiotic and
biotic factors aecting PIMA growth, i.e., changes in basal
area, diameter at breast height, and tree height, shoot elon-
gation, net primary productivity (NPP), and mortality rate by
the end of the 21st century, under the current climate and
predicted climate scenarios. Our intention in emphasising
these eects was to evaluate the impact of climate change
at the individual level, separately from broader processes at
the population, stand, or ecosystem levels such as regenera-
tion, although the latter may be discussed as contextual ele-
ments. Our review was conducted across the current range of
PIMA, which we divided into four main areas (western, cen-
tral, southeastern, and northeastern NA) for analysis. Our spe-
cific objectives were to examine the existing understanding
of how climate change aects the growth and mortality rate
of individual PIMA trees dierently and identify the most-at-
risk areas. We then discuss the combined eects of these fac-
tors on PIMA productivity and mortality rate in the future and
link them to the challenges facing sustainable forest manage-
ment in the future.
Approach
Study area and climatic data
To accurately identify regional climatic trends in NA,
we used the Terrestrial Ecozones of Canada Classification
(Ecological Stratification Working Group 1995), modified by
Price et al. (2013), on which we overlaid the current distribu-
tion of PIMA (Fig. 1). This allowed us to refer to the projec-
tions of Price et al. (2013), which used the same geographi-
cal areas as a baseline for identifying future climate trends
in our study area. We first classified the ecozones into three
distinct groups or study zones, western, central, and east-
Canadian Science Publishing
216 Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075
Fig. 1. Map of the study area across North America. The extent of the four zones was defined by overlaying the Canadian
terrestrial ecozone boundaries (Ecological Stratification Working Group 1995), as modified by Price et al. (2013),totherange
of black spruce across North America. The distributions of discontinuous and continuous permafrost within the black spruce
range are shown in simple and double dashed lines, respectively, based on data from Natural Resources Canada (2022a,2022b).
ern NA, to easily identify the major discrepancies in future
PIMA growth and mortality rates. As eastern NA represents
a more straightforward latitudinal gradient (D’Orangeville et
al. 2016,2018) than their central and western counterparts,
we chose to subdivide this zone into northeastern and south-
eastern NA (Fig. 1). A summary of the main climatic data for
the ecozones studied is presented in Table 1. We finally over-
laid the extent of continuous and discontinuous permafrost
over our study area, based on the data of Natural Resources
Canada (2022a), to easily identify the zones that could be im-
pacted by their respective thaws over the next decades (Fig. 1).
Literature review
We used a twofold filtering process for our literature re-
view. We first conducted an extensive review of the litera-
ture using major electronic databases (e.g., Google Scholar,
Scopus, and Web of ScienceTM) to gather published, peer-
reviewed articles and reports. The search includes three pa-
rameters, with the first being “black spruce” or “Picea mari-
ana”, the second being “climate change” or “temperature” or
“precipitation” or “CO2” and “snow” or “frost” or “wind” or
“soil temperature” or “mineralisation” or “insect” or “com-
petition” or “pathogen” or “parasite” or “disease”, and the
third being “growth” or “mortality”. Due to recent advances
in these areas of forest science in the last few decades, which
were used in our search parameters, we focused primarily
on papers published from 1990 onward. Studies that do not
address the direct relationships between the various biotic
and abiotic factors and PIMA growth or mortality and vari-
ous biotic and abiotic factors were removed from the initial
list. In addition, bibliographies were also scanned to add peer-
reviewed articles that passed our first research filter. The fi-
nal dataset comprised 436 peer-reviewed articles that were
selected as relevant for this literature review.
Climate change-induced abiotic risks to
black spruce growth and mortality rate
Cumulative impacts of temperature and
precipitation
From a physiological point of view, variations in air temper-
ature and precipitation are generally considered as the most
limiting factors for tree growth (Subedi and Sharma 2013).
Depending on the depth of the water table (see Lieers and
Rothwell 1987), PIMA is mostly shallow-rooted and thus has
limited access to deep water resources, making it particularly
sensitive to water stress. It also occurs on drought-sensitive
organic soils, therefore increasing its vulnerability to drought
events (Marchand et al. 2021). While the PIMA range is cur-
rently subject to a double climatic gradient along both the
east–west and north–south axes, the respective influences of
Canadian Science Publishing
Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075 217
Table 1. Summary of the ecozones included in each study zone and their respective climatic parameters, based on data from
Price et al. (2013).
Study zones Ecozones
Mean July (January) air
temperature (◦C)
Mean annual
precipitation (mm)
Mean growing (non-growing)
season precipitation (mm)
Growing season
length (days)
Western NA Taiga Plains 13.8 (−27.2) 347 184 (163) 127
Taiga Cordillera 10.2 (−27.6) 401 172 (229) 94
Boreal Cordillera 10.6 (−21.3) 446 195 (251) 113
Montane Cordillera NA NA NA NA
Central NA Boreal Shield (West) 16.1 (−22.6) 579 371 (208) 159
Boreal Plains 15.4 (−19.2) 472 316 (156) 163
Aspen Parkland 17.0 (−16.6) 451 323 (128) 176
Taiga Shield (West) 12.8 (−29.7) 340 174 (166) 117
Southeastern NA Boreal Shield (East) 14.9 (−16.7) 1000 519 (481) 161
Hudson Plains 13.6 (−24.2) 555 337 (218) 144
Northeastern NA Taiga Shield (East) 10.8 (−22.8) 747 335 (412) 118
temperature and precipitation depend highly on the specific
geographic location being considered.
In eastern Canada, cold temperatures in the northern-
most latitudes exercise a major constraint on PIMA growth,
whereas, further south, increased solar radiation results in
higher evapotranspiration, leading to a moisture-limited en-
vironment (Sniderhan 2018). Currently, the boundary be-
tween temperature- and precipitation-limited environments
in eastern Canada is approximately situated between 49◦N
and 50◦N(D’Orangeville et al. 2016,2018). In contrast, west-
ern and central North America receive about half as much
precipitation as their eastern counterparts (Table 1), result-
ing in an imbalance between evapotranspiration and avail-
able moisture (Price et al. 2013). Central and western NA are
thus almost exclusively moisture-limited until their north-
ernmost extent (Wilmking et al. 2004), and temperature con-
sequently exerts a minor constraint on tree growth in these
regions. Indeed, while low temperatures also largely prevail
in these regions and might intuitively suggest that a large
part of these regions could be temperature-limited, many
studies have shown a negative relationship between tempera-
ture and tree growth in central (Peng et al. 2011) and western
NA (Cahoon et al. 2018). This paradoxical response, often re-
ferred to as the “divergence problem” (see D’Arrigo et al. 2008
for a review), mainly involves temperature-induced drought
stress as the factor explaining reduced tree growth. One ex-
ception is found in the high elevations of central and western
NA (Youngblut and Luckman 2008;Flower and Smith 2011),
where topographic gradients, mainly in the Rocky Mountains
and Alaska Range, lead to the development of stunted or non-
arboreal growth forms resulting from harsh climatic condi-
tions (see Gamache and Payette 2004).
When viewed independently of precipitation (Table 1), the
projected increase in temperatures by 2100 has the poten-
tial to positively aect basal area (Sniderhan et al. 2021)and
height growth (Pau et al. 2022) of PIMA through their ef-
fects on photosynthetic activity (D’Orangeville et al. 2018).
In northeastern NA, Price et al. (2013) predict an increase
in the growing season length by 33 (RCP2.6) to 44 (RCP8.5)
days by 2100 and an increase in mean July air temperature
by 3.1 ◦C (RCP2.6) to 4.8 ◦C (RCP8.5), with increasing precip-
itation expected to fully compensate for higher evapotran-
spiration (climate moisture index (CMI): +5.5 to 8.5 cm un-
der RCP2.6 and RCP8.5, respectively; Table 2). The resulting
enhanced photosynthetic rates north of 50◦N should highly
promote PIMA growth in terms of both height and basal area
(D’Orangeville et al. 2016,2018;Pau et al. 2022), up to +65%
under a +4◦C/+15% precipitation scenario. Model-based stud-
ies estimate a more modest increase in PIMA growth under
a×2CO
2scenario, estimated at a maximum of +25% north of
49◦N(Huang et al. 2013). Payette et al. (1995) and MacDonald
et al. (1998) also suggest a future increase in upright growth
forms at the PIMA treeline in eastern and central Canada
due to increasing temperatures. More importantly, Pau et al.
(2022) demonstrated that even under the RCP8.5 scenario,
the increase in air temperature is not expected to exceed the
threshold at which warming has a negative impact on radial
and height growth in northeastern Canada. In contrast, in-
creasing evapotranspiration (CMI: −1.2 to −3.5 cm; Price et
al. 2013;Table 2) and the strong dependence of autotrophic
respiration on temperatures in southeastern Canada should
be mostly detrimental to PIMA growth but will depend on the
climatic scenario considered (Girardin et al. 2016b;Chaste et
al. 2019). A study conducted by D’Orangeville et al. (2018) in
eastern NA revealed that under a +2◦C temperature increase
associated with a 15% increase in precipitation, enhanced
growth of PIMA is expected over its entire range except for
a thin band covering its southern limit. Conversely, a +4◦C
global increase with associated increased evapotranspiration
would lead to a severe growth decline south of 50◦N, with de-
creases in basal area of up to 45% (D’Orangeville et al. 2018).
Currently, central and western NA receive about half as
much precipitation as their eastern counterparts (Table 1;
Price et al. 2013). Despite the low temperatures prevailing in
these regions, particularly in western NA (Table 1;Price et
al. 2013), most studies revealed that drought is currently the
main factor involved in spruce growth reduction and dieback
episodes across western (Hogg and Bernier 2005) and central
(Kljun et al. 2006) NA. An overall decrease in soil water avail-
ability by 2100 is predicted, with CMI decreases of 8.2 and
Canadian Science Publishing
218 Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075
Table 2. Summary of the projected changes in the selected study zones by 2100 and their individual eects on PIMA growth
and mortality.
Parameters Location Prediction by 2100 Eect on black spruce
Temperature West Increase in MAT by up to 5.5 ◦C (1961–1990
baseline) (Price et al. 2013)
Increase in GSL by 24–46 days (1961–1990
baseline) (Price et al. 2013)
Increase in basal area (Sniderhan et al. 2021) and height
growth (Pau et al. 2022), but constrained by moisture (Pau
et al. 2022)
Centre Increase in MAT by 5.5 ◦C (1961–1990 baseline)
(Price et al. 2013)
Increase in GSL by 21–35 days (1961–1990
baseline) (Price et al. 2013)
Increase in basal area (Sniderhan et al. 2021) and height
growth (Pau et al. 2022), but constrained by moisture (Pau
et al. 2022)
South-east Increase in MAT by 5.5◦C (1961–1990 baseline)
(Price et al. 2013)
Increase in GSL by 31–43 days (1961–1990
baseline) (Price et al. 2013)
Increase in basal area (Sniderhan et al. 2021) and height
growth (Pau et al. 2022) due to longer growing season, but
constrained by moisture (Pau et al. 2022)
North-east Increase in MAT by 3.5–6 ◦C (1961–1990
baseline) (Price et al. 2013)
Increase in GSL by 33–44 days (1961–1990
baseline) (Price et al. 2013)
Increase in height (Gamache and Payette 2004;Pau et al.
2022), basal area growth (Silva et al. 2010;D’Orangeville et
al. 2016), and leader shoot elongation (Gamache and
Payette 2004) due to a currently temperature-limited
environment associated to longer growing season
(D’Orangeville et al. 2016;Sniderhan et al. 2021;Pau et al.
2022)
Precipitation and
water availability
West Increase in MAP by 10%–25% (Price et al. 2013)
CMI: +2.8 to −1.4 cm (1961–1990 baseline)
(Price et al. 2013)
Decrease in general PIMA growth (D’Orangeville et al. 2016),
net primary productivity (Ma et al. 2012), and basal area
growth, up to 25% (Ma et al. 2012;D’Orangeville et al.
2018;Sniderhan et al. 2021), independently of elevation
gradients (Walker and Johnstone 2014;Walker et al. 2015)
Centre Increase in MAP by 11% (Price et al. 2013)
CMI: −2.4 to −8.2 cm (1961–1990 baseline)
(Price et al. 2013)
Decrease in general PIMA growth (D’Orangeville et al. 2016),
net primary productivity (Ma et al. 2012), and basal area
growth, up to 25% (Ma et al. 2012;D’Orangeville et al.
2018;Sniderhan et al. 2021), independently of elevation
gradients (Walker and Johnstone 2014;Walker et al. 2015)
South-east Increase in MAP by 13% (Price et al. 2013)
CMI: −1.2 to −3.5 cm (1961–1990 baseline)
(Price et al. 2013)
Decrease in general PIMA growth (D’Orangeville et al. 2016),
net primary productivity (Ma et al. 2012), and basal area
growth, up to 45% (Ma et al. 2012;D’Orangeville et al.
2018)
North-east Increase in MAP by 10%–22% (Price et al. 2013)
CMI: +5.5–8.5 cm (1961–1990 baseline) (Price
et al. 2013)
Increase in general PIMA (D’Orangeville et al. 2016)and
basal area growth, from +25% (Huang et al. 2013)toup
to +65% (D’Orangeville et al. 2018)
Atmospheric CO2
concentration
West 421 ppm (RCP2.6) to 936 ppm (RCP8.5) (Pörtner
et al. 2022)
Theoretical increase in net primary productivity by 23%–55%
(Johnsen and Seiler 1996;Bigras and Bertrand 2006), and
radial and height growth (Messaoud and Chen 2011), but
may be more moderate in reality (Gedalof and Berg 2010;
Girardin et al. 2016a)
Centre 421 ppm (RCP2.6) to 936 ppm (RCP8.5) (Pörtner
et al. 2022)
Theoretical increase in net primary productivity by 23%–55%
(Johnsen and Seiler 1996;Bigras and Bertrand 2006), and
radial and height growth (Messaoud and Chen 2011), but
may be more moderate in reality (Gedalof and Berg 2010;
Girardin et al. 2016a)
South-east 421 ppm (RCP2.6) to 936 ppm (RCP8.5) (Pörtner
et al. 2022)
Theoretical increase in net primary productivity by 23%–55%
(Johnsen and Seiler 1996;Bigras and Bertrand 2006), and
radial and height growth (Messaoud and Chen 2011), but
may be more moderate in reality (Gedalof and Berg 2010;
Girardin et al. 2016a)
North-east 421 ppm (RCP2.6) to 936 ppm (RCP8.5) (Pörtner
et al. 2022)
Theoretical increase in net primary productivity by 23%–55%
(Johnsen and Seiler 1996;Bigras and Bertrand 2006), and
radial and height growth (Messaoud and Chen 2011), but
may be more moderate in reality (Gedalof and Berg 2010;
Girardin et al. 2016a)
Snow West Increase in snow cover by 1%–50% (1961–1990
baseline) (Brown and Mote 2009)
Increase in soil insulation during winter or
no change (Groman et al. 2001)
Probable decrease in frost damages at the root level
(Groman et al. 2001), but increase on the apical
apparatus (Ma et al. 2019).
But poorly studied in NA boreal forests (Fréchette et al.
2011). More research is needed
Canadian Science Publishing
Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075 219
Table 2. (continued).
Parameters Location Prediction by 2100 Eect on black spruce
Centre Decrease in snow cover by 1%– 50% (1961–1990
baseline) (Brown and Mote 2009)
Decrease in soil insulation during winter
(Groman et al. 2001)
Probable increase in frost damages at the root level
(Groman et al. 2001) and on the apical apparatus (Ma et
al. 2019).
But poorly studied in NA boreal forests (Fréchette et al.
2011). More research is needed
South-east Decrease in snow cover by 1%–10% (1961–1990
baseline) (Brown and Mote 2009)
Decrease in soil insulation during winter
(Groman et al. 2001)
Probable increase in frost damages at the root level
(Groman et al. 2001) and the apical apparatus (Ma et al.
2019)
Decreased net primary productivity of PIMA stands
(Fréchette et al. 2011)
But poorly studied in NA boreal forests (Fréchette et al.
2011). More research is needed
North-east Stable snow cover or increase by up to 10%
(1961–1990 baseline) (Brown and Mote 2009)
Increase in soil insulation during winter or
no change (Groman et al. 2001)
Probable decrease in frost damages at the root level
(Groman et al. 2001), but increase on the apical
apparatus (Ma et al. 2019).
But poorly studied in NA boreal forests (Fréchette et al.
2011). More research is needed
Frost West Decrease in snow crystal abrasion
Decreasing probability of frost damage
(Nitschke and Innes 2008)
40% probably of decreasing frost damage during growing
season (early 21st century baseline) (Nitschke and Innes,
2008)
Centre Increase in snow crystal abrasion
Increased frost damage during growing
season (Liu et al. 2018)
Increase in fine root mortality (Groman et al. 2001)
Increase in buds and shoots damage (Marquis et al. 2022)
South-east Increase in snow crystal abrasion
Increased frost damage during growing
season (Groman et al. 2001)
Increase in fine root mortality (Groman et al. 2001)
Increase in buds and shoots damage (Marquis et al. 2022)
North-east Decrease in snow crystal abrasion (Gamache
and Payette 2005)
Increase in height growth (Gamache and Payette 2005)
Edaphic
parameters
West Increased soil temperature (Dao et al. 2015)
Increased nitrogen mineralisation and
nitrification (Rustad et al. 2001)in
permafrost-free areas
Large-scale permafrost thaw (Wisser et al.
2011)
Increased tree growth in permafrost-free areas (Melillo et al.
2011)
Increase in freezing injuries due to earlier budbreak (Li et
al. 2015)
Mortality episodes in areas underlain by permafrost
(Baltzer et al. 2014)
Centre Increased soil temperature (Dao et al. 2015)
Increased nitrogen mineralisation and
nitrification (Rustad et al. 2001)in
permafrost-free areas
Large-scale permafrost thaw (Wisser et al.
2011)
Increased tree growth in permafrost-free areas (Melillo et al.
2011)
Increase in freezing injuries due to earlier budbreak (Li et
al. 2015)
Mortality episodes in areas underlain by permafrost
(Camill et al. 2001)
South-east Increased soil temperature (Dao et al. 2015)by
1.9–3.3 ◦C by 2080 (Houle et al. 2012)
Increased nitrogen mineralisation and
nitrification (Rustad et al. 2001)in
permafrost-free areas
Increased tree growth (Melillo et al. 2011)
Increase in freezing injuries due to earlier budbreak (Li et
al. 2015)
Mortality episodes in areas underlain by permafrost
(Pelletier et al. 2019)
North-east Increased soil temperature (Dao et al. 2015)
Increased nitrogen mineralisation and
nitrification (Rustad et al. 2001)in
permafrost-free areas
Large-scale permafrost thaw (Wisser et al.
2011)
Increased tree growth in permafrost-free areas (Melillo et al.
2011)
Increase in freezing injuries due to earlier budbreak (Li et
al. 2015)
Mortality episodes in areas underlain by permafrost
(Pelletier et al. 2019)
ESB outbreaks West Reduced defoliation at the southern limit of
ESB distribution, but expansion northward
and at higher elevations (Régnière et al. 2012)
Increased severity and duration of outbreaks
on PIMA (Fuentealba et al. 2017;
Bellemin-Noël et al. 2021)
Increased phenological synchrony between ESB and PIMA
budburst (Bronson et al. 2009;Bellemin-Noël et al. 2021)
Increased PIMA growth reduction and mortality episodes
in high latitudes (Pureswaran et al. 2015) and altitudes
(Régnière et al. 2012)
Centre Reduced defoliation at the southern limit of
ESB distribution, but expansion northward
and at higher elevations (Régnière et al. 2012)
Increased severity and duration of outbreaks
on PIMA (Fuentealba et al. 2017;
Bellemin-Noël et al. 2021)
Increased phenological synchrony between ESB and PIMA
budburst (Bronson et al. 2009;Bellemin-Noël et al. 2021)
Increased PIMA growth reduction and mortality episodes
in high latitudes (Pureswaran et al. 2015)
Canadian Science Publishing
220 Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075
Table 2. (concluded).
Parameters Location Prediction by 2100 Eect on black spruce
South-east Reduced impact at the southern limit of ESB
distribution (Régnière et al. 2012)
Increased phenological synchrony between ESB and PIMA
budburst (Bronson et al. 2009;Bellemin-Noël et al. 2021)
Increased PIMA growth reduction and mortality episodes,
but decrease at the southernmost limit (Fuentealba et al.
2017;Bellemin-Noël et al. 2021)
North-east Increased severity of outbreaks at the northern
limit of ESB distribution, but may be
modulated by tree density (Gray 2008;
Régnière et al. 2012)
Increased duration of outbreaks at the
northern limit of ESB distribution (Gray 2008)
Increased phenological synchrony between ESB and PIMA
budburst (Bronson et al. 2009;Bellemin-Noël et al. 2021)
Increased PIMA growth reduction and mortality episodes
(Fuentealba et al. 2017;Bellemin-Noël et al. 2021)
Competition West Increased proportion of broadleaf (mainly
Populus and Betula)taxa(Walker et al. 2017;
Mack et al. 2021)
Increased proportion of white spruce in areas
subjected to permafrost thaw (Nicklen et al.
2021)
Changes in interspecific competition, increased radial and
height growth (Légaré et al. 2004;Chavardès et al. 2021,
2023). Increased interspecific competition in areas
subjected to permafrost thaw (Trugman et al. 2018).
More research is needed
Centre Increased proportion of broadleaf (mainly
Populus and Betula)taxa(Walker et al. 2017;
Mack et al. 2021)
Increased proportion of white spruce in areas
subjected to permafrost thaw (Nicklen et al.
2021)
Changes in interspecific competition, increased radial and
height growth (Légaré et al. 2004;Chavardès et al. 2021,
2023). Increased interspecific competition in areas
subjected to permafrost thaw (Trugman et al. 2018)
More research is needed
South-east Increasing number of regeneration failures
(Molina et al. 2021), or replacement by jack
pine (Boulanger et al. 2017;Baltzer et al. 2021)
Reduced intraspecific competition, increased PIMA growth
More research is needed
North-east Increased proportion of jack pine (Baltzer et al.
2021)
More research is needed
Diseases and
parasites
West Northward expansion of diseases and parasites
(Kliejunas et al. 2009)
Increased impact of diseases and parasites
through increased drought stress
Increased growth reduction and mortality rates in areas
infected by diseases and parasites
More research is needed
Centre Northward expansion of ESDM and more
generally of diseases and parasites (Kliejunas
et al. 2009)
Increase by 200% of the area infected by
ESDM (Westwood et al. 2012)
Increased impact of diseases and parasites
through increased drought stress
Increased growth reductions and mortality rates in areas
infected by ESDM (Westwood et al. 2012) or other diseases
and parasites
More research is needed
South-east Northward expansion of ESDM and more
generally of diseases and parasites (Kliejunas
et al. 2009)
Increased growth reduction and mortality rates in areas
infected by ESDM (Westwood et al. 2012) or other diseases
and parasites
More research is needed
North-east Northward expansion of ESDM and more
generally of diseases and parasites (Kliejunas
et al. 2009)
Increased growth reduction and mortality rates in areas
infected by ESDM (Westwood et al. 2012) or other diseases
and parasites
More research is needed
Note: MAT, mean annual temperatures; MAP, mean annual precipitation; GSL, growing season length; WAI, water availability index; CMI, climate moisture index.
1.4 cm being estimated for central and western NA, respec-
tively (Table 2;Price et al. 2013). These regions should thus
experience the strongest increase in the frequency and in-
tensity of drought events across the entire PIMA range (Price
et al. 2013;Tam et al. 2018). This should lead to severe de-
clines in black spruce NPP (Girardin et al. 2008;Ma et al.
2012;Girardin et al. 2016b), reaching up to 25% under the
RCP8.5 scenario by the end of the century (Girardin et al.
2016b). It seems important to note that western NA also pos-
sesses significant topographic gradients due to the presence
of mountainous areas, particularly in the Rocky Mountains
and Alaska. These high-altitude regions are dominated by
wetter and colder conditions, thus suggesting a control of
PIMA growth by temperature and consequently a positive
future response to climate change (Hogg and Bernier 2005;
Walker and Johnstone 2014). Surprisingly, studies conducted
in the interior of the Alaska Range and in the northern Yukon
revealed that the PIMA growth response to increased tem-
perature was negative and largely independent of latitudinal
or elevation gradients (Walker and Johnstone 2014;Walker
et al. 2015). These studies suggest that PIMA in central and
western North America will become increasingly susceptible
to drought stress by the end of the century, independent of
temperature gradients related to altitude or elevation.
Canadian Science Publishing
Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075 221
CO2concentration
CO2concentrations have increased from 310 ppm in 1950
to 421 ppm in 2023, mainly due to the burning of fossil fuels
by human activities. According to current projections, these
emissions may remain stable (RCP2.6), increase moderately to
670 ppm (RCP6), or, in the most pessimistic scenario, reach
up to 936 ppm (RCP8.5) by the year 2100 (Table 2;Pörtner
et al. 2022). Under the present climate policies, recent pro-
jections suggest that the latter two are currently the most
likely (Capellán-Pérez et al. 2016), thus suggesting high at-
mospheric [CO2] enrichment by 2100. As photosynthetic up-
takes of carbon-driving NPP are not saturated at current at-
mospheric levels (Long et al. 2004), CO2fertilisation may have
a net positive eect on PIMA growth. At a physiological level,
it is predicted that elevated [CO2] would allow for better net
carbon fixation and shade tolerance through more ecient
use of light energy (Drake et al. 1997). In this sense, CO2-
enrichment experiments realised under a 710 ppm [CO2]sce-
nario increased the NPP, height, diameter, shoot, and root dry
mass of PIMA seedlings by 23% (Bigras and Bertrand 2006),
while similar experiments suggest a 55% NPP increase under
the same atmospheric conditions (Johnsen and Seiler 1996).
Other theoretical experiments on PIMA seedlings also sug-
gest an increase in PIMA growth over its entire range as a
consequence of rising [CO2] as compared to the preindustrial
era (Table 1;Li et al. 2015).
However, several studies using tree rings have not been
able to objectively show this increase in tree growth at the
global scale (Gedalof and Berg 2010) and more specifically
for boreal species (Price et al. 2013;Girardin et al. 2016a). In-
deed, the response of mature trees may dier from that of
seedlings, but it is also more challenging to study due to nu-
merous confounding factors. As a result, our understanding
of the response of mature PIMA trees compared to seedlings
remains limited at present. Recent tree-ring studies over the
last 50 years have struggled to identify any significant impact
of increased atmospheric [CO2] levels, mainly due to these
confounding factors and the relatively modest rise in [CO2]
levels (Gedalof and Berg 2010), while a study conducted in
British Columbia by Messaoud and Chen (2011) on mature
PIMA trees suggests a positive response to recent increases
in [CO2]. Free-air CO2enrichment (FACE) experiments have
the potential to improve our knowledge of mature trees but
have so far rarely been carried out on mature trees, and to
our knowledge, not on PIMA. In addition to the direct impact
of [CO2] on the growth of PIMA, Bigras and Bertrand (2006)
showed that higher atmospheric [CO2] was related to earlier
bud set in the fall. As an earlier bud set reduces the likelihood
of damage from early frost events, [CO2] enrichment also has
the potential to decrease frost damage to PIMA buds in the
upcoming decades.
Interactions between snow and frost
Boreal trees have naturally evolved to cope with intense
winter stress, characterised by long frost periods and sub-
stantial snow accumulation that largely aect their phenol-
ogy. The decreasing photoperiod and reduced temperatures
in fall induce the development of buds and enable cold hardi-
ness, while warmer temperatures and the longer photoperiod
in spring initiate budburst and trigger annual shoot growth
(Chang et al. 2021). Under a warming climate, the photope-
riod will remain unchanged while predicted temperatures
will rise, resulting in an increasing mismatch between these
two parameters that may happen more frequently. This will
modify the phenological cycles of trees, making them in-
creasingly vulnerable to future freezing conditions. Indeed,
earlier dehardening of PIMA buds followed by below-freezing
temperatures can induce injuries to meristems, leaves, or
roots in spring (Man et al. 2021). Early frost conditions in
the fall may have similar impacts or create xylem cavita-
tion (Sperry and Robson 2001), ultimately leading to reduced
PIMA growth. In addition to these impacts on the apical ap-
paratus, prolonged ground freezes can also damage the fine
roots of seedlings or mature trees. When there is insucient
snow cover in winter, roots are particularly vulnerable, as
snow acts as an insulator and protects PIMA roots against
frost in spring and fall (Fréchette et al. 2011). Consequently,
frost can largely hinder PIMA growth, cause permanent dam-
age, and even lead to individual death several years following
a frost event (Gaumont-Guay et al. 2008).
Winter conditions as a determinant of tree growth have
often been overlooked during the past decades, despite
their major impact in northern environments (Gamache and
Payette 2004), due to the uncertainties related to future win-
ter conditions. Snow and frost responses to global warming
will indeed vary greatly with latitude, longitude, elevation,
and seasonality, inducing complex predictions for their re-
spective influences on PIMA growth. For example, the model
of Brown and Mote (2009) predicts a large decrease in the
snow cover by the year 2070 south of 60◦N in western and cen-
tral NA, while in eastern NA this transition occurs between
51 and 53◦N(Table 2). Snow cover is also predicted to sig-
nificantly decrease in the coastal areas of western and east-
ern NA, and, conversely, increase north of the Great Lakes
(Table 2;Brown and Mote 2009).
As snow is a significant ground insulator, several hypothe-
ses have been suggested for how predicted changes will aect
PIMA growth. Firstly, freeze events are likely to become in-
creasingly frequent in the NA regions, where significant de-
creases in snow cover are predicted (Groman et al. 2001).
Root damage is thus more likely to occur and may induce re-
ductions in PIMA growth or mortality episodes. In contrast,
the potential for root damage is not likely to be dierent
from current conditions in regions that are expected to ex-
perience an increase in snow cover. Secondly, higher tem-
peratures and longer growing seasons across the entire PIMA
range should induce an earlier emergence from winter dor-
mancy, during which sub-zero temperatures are still com-
mon (Ma et al. 2019), but at a time when new needles have
low frost resistance. Even though PIMA is highly adapted to
below-freezing temperatures (Girardin et al. 2022), the aver-
age number of frost days during the growing season has in-
creased by more than two weeks between 1982 and 2012, a
tendency that is likely to continue over the next decades (Liu
et al. 2018). Since PIMA is an early leaf-out species (Girardin
et al. 2022), we can anticipate increased damage to the buds
and shoots of both mature trees and seedlings due to climate
Canadian Science Publishing
222 Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075
change, likely leading to significant reductions in individual
growth (Groman et al. 2001;Marquis et al. 2022).
Currently, harsh winter conditions, characterised by thick
snowpatches, frost desiccation of needles and buds, and
mechanical damage associated with high wind speeds, pre-
vail towards the forest-tundra ecotone (Truchon-Savard et
al. 2019;Maher et al. 2020). These conditions hinder the
development of erect PIMA forms due to damage caused
by snow and ice, resulting in the development of stunted
(“krummholz”) growth forms (Truchon-Savard et al. 2019).
In the next few decades, climate change in the northern
range of PIMA is expected to result in an amelioration of cli-
matic conditions through reduced ice crystal abrasion and
thicker snow cover during the winter (Brown and Mote 2009),
thus resulting in significant growth gains at the individual
level. An increased development of upright PIMA growth
forms has already been identified at the treeline in north-
ern Quebec in response to 20th-century global warming, a
tendency that is likely to continue in the future with in-
creasing mean annual temperatures (Gamache and Payette
2004).
Edaphic parameters
PIMA forest ecosystems are largely dominated by cold,
nutrient-poor, permafrost-dominated soils, which induce an
overall low NPP (Cleve et al. 1993). Combined with short
growing seasons and limited decomposition rates due to cold
air temperatures, it partly explains why boreal forests store
important quantities of biogenic carbon, and play a major
role in climate regulation through the carbon cycle (Snyder
et al. 2004). Soil warming due to climate change and in-
creased wildfire activity may aect many chemical processes
in forested ecosystems and is also expected to stimulate per-
mafrost thawing, ultimately leading to major changes in
PIMA growth and mortality rates.
Most of the boreal zone is currently subject to nitrogen
limitations (Norby et al. 2010). Soil temperature is indeed a
key factor aecting processes such as litter decomposition,
soil respiration, nutrient acquisition by plant species, and
fine root growth dynamics (Rustad et al. 2001). At a phys-
iological level, high soil temperatures improve xylem ac-
tivity and water uptake of gymnosperms (Alvarez-Uria and
Körner 2007) and are also positively correlated to the de-
composition and mineralisation rates of organic materials,
themselves driven by temperature-limited microbial activity
(Rustad et al. 2001;Lafleur et al. 2015). Consequently, the
predicted increase in soil temperatures over the upcoming
decades is expected to stimulate the overall growth of bo-
real trees. This will occur through the enhancement of ni-
trogen mineralisation and nitrification processes in organic
soils (Campbell et al. 2009;Price et al. 2013). More partic-
ularly, Moore et al. (1999) predict a 4–7% increase in litter
decomposition rates under a 500–700 ppm [CO2] scenario,
which could stimulate PIMA growth across its entire range.
However, few studies have been carried out on the specific re-
sponse of spruce compared with other conifer species. Peng
and Dang (2003) and Dang and Cheng (2004) have shown from
soil warming experiments conducted on PIMA seedlings that
its optimum soil temperature for total biomass is around 16.0
◦C. When summer soil temperatures reach approximately
15 ◦C in southeastern NA (Lupi et al. 2012), with projected
increases between 1.9 and 3.3 ◦C by the year 2080 (Houle
et al. 2012), PIMA productivity across its entire range is ex-
pected to be positively impacted, as the temperatures will
be reaching a beneficial threshold in an extended part of
the year. Moreover, a study conducted in eastern Canada by
Lafleur et al. (2015) suggests a decrease in peat production
rates by the end of the century under an increasing frequency
of high-severity fires. The related decrease in peat produc-
tion rates may also reduce the area covered by paludified
forests in southeastern NA, thus increasing the general pro-
ductivity of eastern Canadian forests in permafrost-free ar-
eas.
With the disappearance of permafrost expected within the
next 50–100 years (Price et al. 2013), large osets of posi-
tive eects related to warmer soils can also be expected. Cur-
rently, most of central and western North America is cov-
ered by discontinuous permafrost, although continuous per-
mafrost covers a thin strip north of the PIMA range (Fig. 1).
In eastern NA, discontinuous permafrost covers the entire
region above 52◦N(Fig. 1) and is often surrounded by satu-
rated, permafrost-free wetlands with few trees (Zhang et al.
2003). Currently, peat plateaus cover between 30% and 70%
of peatlands in the discontinuous permafrost zone, where
PIMA is the more common species (Olefeldt et al. 2021). Due
to warming soils, permafrost was estimated to thaw by ap-
proximately 0.58% per year between 2000 and 2015 in west-
ern Canada, a tendency that is predicted to continue in the
upcoming decades (Chasmer and Hopkinson 2017). The re-
sulting deepening of the active layer will lead to major hydro-
logical changes, including increased waterlogged conditions
due to the transformation of discontinuous permafrost into
thermokarst (Olefeldt et al. 2021), to which PIMA is poorly
adapted (Islam and Macdonald 2004). At the physiological
level, flooding conditions impair root functions and water
uptake; thus, reduced PIMA growth could be largely expected
in these areas (Baltzer et al. 2014). Moreover, the intolerance
of PIMA to flooding (Islam and Macdonald 2004) and the in-
creased transformation of dry peat plateaus into inundated
wetlands should result in large-scale PIMA mortality episodes
in the next decades (Haynes et al. 2021). This negative ef-
fect could even be amplified by the increasing soil instabil-
ity due to permafrost thaw (Van Cleve et al. 1990), resulting
in an increased PIMA mortality rate across the current per-
mafrost zone. Observations of this eect are already visible in
the eastern (Pelletier et al. 2019), central (Camill et al. 2001),
and western NA (Baltzer et al. 2014). Additionally, recent stud-
ies have shown that wildfire activity is tightly linked to per-
mafrost thaw, in part due to the low albedo of the burned
zone, which leads to higher soil temperatures. About 25% of
the peat plateaus of western Canada have burned in the last
30 years (Gibson et al. 2018), and this tendency is predicted
to increase in the next decades, particularly in central and
western NA (de Groot et al. 2013). Regions where wildfire ac-
tivity is predicted to increase could thus become increasingly
vulnerable to permafrost thaw in the next decades (Turetsky
et al. 2011).
Canadian Science Publishing
Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075 223
Climate change-induced biotic risks
Eastern spruce budworm
Insect outbreaks are, with fire, among the most severe eco-
logical disturbances in NA boreal forests (Jardon et al. 2003).
Among these, the eastern spruce budworm (Choristoneura fu-
miferana (Clem.), hereafter ESB), a conifer-defoliating moth
widely distributed across the PIMA range, is considered the
main insect pest for PIMA (Fierravanti et al. 2015). It feeds
on the young needles of fir (Abies spp.) and spruce (Picea spp.),
inducing tree damage ranging from simple annual growth re-
duction to the death of large stands after repeated major de-
foliation years (Chen et al. 2017). As a flush feeder, ESB larvae
begin to feed on the newest annual foliage. The phenology be-
tween larvae emergence from diapause and the timing of the
budburst of their hosts thus highly determines their specific
impact. Historically, balsam fir (Abies balsamea (L.) Mill.) bud-
burst occurred a few days after the emergence of the second
ESB instar from winter diapause and on average 13 days later
for PIMA (Nealis and Régnière 2004). While PIMA remains
an attractive host in terms of nutritional value for the ESB
(Pureswaran et al. 2015), this phenological asynchrony largely
explains the reduced impact of defoliation on this species
during 20th-century outbreaks. However, as insect species are
extremely sensitive to changes in spring temperatures, many
defoliating insects show an advanced phenology in response
to temperature increases (Pureswaran et al. 2015;Fuentealba
et al. 2017;Bellemin-Noël et al. 2021). While an even ear-
lier emergence of ESB could increase the phenological lag
for PIMA budburst, rising temperatures in boreal ecosystems
also aect tree species phenology (Wolkovich et al. 2012). Cur-
rently, field experiments (see Bronson et al. 2009;Bellemin-
Noël et al. 2021) have shown that warmer temperatures will
probably improve ESB–PIMA phenological synchrony and in-
crease caterpillar growth rates. As a result, PIMA could ex-
perience defoliation rates close to those of balsam fir in the
next decades (Fuentealba et al. 2017;Bellemin-Noël et al.
2021), although defoliation rates may also be moderated by
the increase in mixed stands due to a growing impact of ESB
(Cappuccino et al. 1998).
While its host range currently extends beyond the 60th
parallel, the northern limit of the ESB in eastern NA rarely
reaches the ecotone between closed-crown forest and open-
canopy woodlands, which is situated at approximately 52◦N
(Gray 2008). Despite the presence of their hosts, short sum-
mers in the high latitudes prevent ESB larvae from complet-
ing their development cycle and limit their northern distri-
bution. Thus, the spatial variability of ESB distribution is not
only related to forest composition but also depends on the
climatic conditions necessary for its development, particu-
larly temperature. In eastern Canada, recent climatic mod-
els suggest a reduction in the prevalence of epidemics at the
southernmost limit of ESB distribution (Régnière et al. 2012;
Pureswaran et al. 2019). Conversely, the northernmost ESB
populations will no longer be limited by summers too short
to complete their life cycle. In consequence, an extension of
the northern limit of defoliation and a shift to higher ele-
vations are likely to happen (Gray 2008;Pureswaran et al.
2015;Régnière et al. 2012). As ESB population dynamics are
impacted by many climatic, biological, and physical factors,
its future extension to the north remains uncertain. For ex-
ample, Gray (2008) suggests an increase in the severity and
duration of outbreaks northwards by the end of the century.
On the other hand, Régnière et al. (2012) remain less skep-
tical and suggest that the low density of PIMA in the lichen
woodlands of eastern Canada should prevent the increasing
prevalence of outbreaks until the year 2100. In this sense,
further studies (see Cappuccino et al. 1998)havealsoshown
that the climate-induced northward shift in ESB epidemics
could be moderated by the predicted increase in mixed
stands.
Similar to their eastern counterparts, the model of
Régnière et al. (2012) suggests a decreasing prevalence of out-
breaks in the southern part of western and central NA due
to unsuitable climatic conditions. However, higher densities
of white (Picea glauca (Moench) Voss) and black spruce cur-
rently prevail at the northernmost latitudes in these regions.
Consequently, this model predicts that the higher density of
ESB’s host plants in northern latitudes may largely enhance
the prevalence of outbreaks in these regions due to climate
change. However, to our knowledge, the specific response of
PIMA has not been studied in these regions, and the ques-
tion of the duration of ESB outbreaks remains uncertain.
While Price et al. (2013) suggest longer epidemics in the fu-
ture, the models of Boulanger et al. (2016) in central and east-
ern Canada suggest little change in their duration under the
RCP2.6 scenario, while the RCP8.5 would induce a shortening
of epidemic duration. At the same time, ESB epidemics are
expected to develop at higher altitudes in the mountainous
regions of western Canada, where harsh conditions currently
limit their spatial distribution (Régnière et al. 2012). This in-
creased impact of ESB in western and central NA may also be
moderated by a rise in the proportion of deciduous species in
western and central NA stands, caused both by the increas-
ing frequency, size, and severity of fires and insect outbreaks
(Cappuccino et al. 1998).
Competition
Competitive relationships for water, nutrients, or light
have been shown to exert a major impact on PIMA growth
(Légaré et al. 2004;Montoro Girona et al. 2017). More specifi-
cally, recent studies (e.g., Ettinger and HilleRisLambers 2013;
Oboite and Comeau 2020) demonstrated that competition
may have a greater impact than climate on the radial and
height growth of tree species, making it one of the most in-
fluential natural processes in NA boreal forests. PIMA growth
and mortality rate indeed depend not only on climate or
disturbance regimes but also on the surrounding environ-
ment, including species diversity. As ecosystem dynamics
are shaped by patterns and processes of disturbance and re-
covery, particularly fire, they are likely to be largely mod-
ified under predicted climate change. PIMA responses may
thus strongly depend on future system states (Trugman et
al. 2018), including resilience-driven ecosystem composition.
However, strong regional dierences in the recovery patterns
of PIMA are expected (Baltzer et al. 2021), which may modify
their competitive relationships.
Canadian Science Publishing
224 Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075
Although they are largely dependent on future climate sce-
narios, large-scale studies conducted in southeastern Canada
show that future changes in fire regimes (+33% in fire oc-
currence (Wotton et al. 2010)and+39% to 157% in area
burned (Price et al. 2013) by the year 2100) and intensify-
ing forest management are expected to aect the regener-
ation potential of PIMA and lead to an increasing number
of regeneration failures (Boulanger et al. 2017;Molina et al.
2021;Baltzer et al. 2021). Through reversion from closed-
crown forest to open lichen woodland, the decrease in gen-
eral PIMA stand density (Splawinski et al. 2019a) should re-
duce intraspecific competition, particularly for ground water
(Ameray et al. 2023), and conversely largely promote PIMA
growth at the individual level (Oboite and Comeau 2020).
For example, the models of Augustin et al. (2022) state that
more than 30% of southeastern Canada will be covered by
open woodlands under the RCP4.5 and RCP8.5 scenarios by
the year 2100. On the other hand, a transcontinental-scale
study conducted by Baltzer et al. (2021) suggests that replace-
ment of PIMA by jack pine (Pinus banksiana Lamb.) is the
most probable outcome in northeastern and southeastern
NA.
Dierent scenarios are forecast for central and western
NA. Due to shorter fire cycles (typically <100 years) and a
predicted larger burned area (up to +124% in western NA
and +143% in central NA by 2100, Price et al. 2013), it is pre-
dicted that pure PIMA stands will shift to a deciduous dom-
inance in western (Johnstone et al. 2010;Walker et al. 2017;
Mack et al. 2021) and central NA (Baltzer et al. 2021). More
specifically, mixed PIMA-broadleaves stands dominated by as-
pen (Populus spp.)andbirch(Betula spp.) should dominate es-
pecially on moderately to well-drained sites (Johnstone et al.
2010). Studies conducted in eastern Canada (see Légaré et al.
2004;Chavardès et al. 2021,2023) pointed out that aspen
tends to increase the nutrient mineralisation rate in mixed
aspen-PIMA stands, thus increasing radial and height growth
of the latter. Positive eects on PIMA related to interspecific
relationships are thus likely to occur in western and central
NA.
Superimposed on these longitudinal gradients,
permafrost-thaw could also induce major changes in PIMA
competitive relationships. For example, Nicklen et al. (2021)
emphasised that the faster-growing and deeper-rooted white
spruce may outcompete PIMA in areas currently underlain
by permafrost through better use of the active layer under
permafrost thaw, resulting in increased drought stress for
PIMA. In consequence, interspecific competition should be
largely detrimental to PIMA in regions currently underlain
by permafrost (Fig. 1)(Trugman et al. 2018). However, intra-
and interspecific competition mechanisms linked to PIMA
remain poorly investigated in boreal forest (Chavardès et al.
2023) and are intrinsic to complex interactions with other
disturbance factors. Further research is required to better
assess their impacts on individual PIMA growth.
Diseases and parasites
Diseases and parasites are major components of forest
ecosystems, and thus have significant consequences that
range from individual tree growth to stand-level structure,
composition, and dynamics (Skay et al. 2021). Currently,
about 20 species of pathogens endemic to North America
attack PIMA (Canadian Forest Service 2010). For example,
fungi belonging to the genus Armillaria (Fr.) Staude infect
over 200 million hectares across Canada (Canadian Forest Ser-
vice 2010), with an estimated 116 000 m3of PIMA growth
loss per year in the Canadian province of Manitoba (Brandt
1995). Another example is eastern dwarf mistletoe (Arceutho-
bium pusillum Peck, hereafter ESDM), which is a parasitic vas-
cular plant on conifers that extends across eastern and cen-
tral NA, causing large growth declines in infested trees. After
about 15 years, 75% of trees succumb to the eects of the
parasite (Ostry and Nicholls 1979). Despite their major im-
pact on stand dynamics, the interaction between pathogens,
parasites, and climate change has surprisingly received little
attention in boreal forests. Consequently, their future evolu-
tion remains highly uncertain, and further research is def-
initely needed. As climate change aects pathogens, hosts,
and their interactions, various general hypotheses can, how-
ever, be put forward, although their longitudinal and latitu-
dinal variability remain dicult to interpret.
First, climatic constraints currently limit the range of many
diseases and parasites, such as ESDM (Kliejunas et al. 2009).
Longer growing seasons and a decreasing annual duration
of snow or frost in the soil will likely result in an exten-
sion of their range to higher latitudes and elevations, thus
exacerbating their impact on PIMA in regions where they
were previously limited (Gray et al. 2021). Second, infec-
tion potential and the spread of diseases and parasites are
highly related to tree stress (Zhang and Sutton 2011). An
increasing frequency of drought events, particularly over
the central and western parts of the PIMA range, may re-
sult in increasingly stressful conditions, thus enhancing the
impact of diseases and parasites on its growth and mor-
tality rate in these regions. These stressful conditions are
likely to be reduced in the northeastern regions of Canada,
where moisture is less limited. However, Kliejunas et al.
(2009) remind us that increasing precipitation in spring
may consequently increase the impact of foliar diseases,
while enhanced precipitation during the fall may promote
PIMA stem rusts. Finally, Westwood et al. (2012) propose
that the area infected by Armillaria spp. could triple in the
next 50 years under climate change scenarios in central
Canada, suggesting an increasing impact of these fungi on
PIMA.
Complex relationships between factors
and their link with ecosystem-scale
disturbances
Our review highlights the importance of available mois-
ture on the growth and mortality rate of PIMA, which will
largely vary over its entire range. These changes will be par-
ticularly striking in western and central NA, where they are
predicted to translate into sharp decreases in PIMA growth
and abrupt increases in mortality episodes by 2100 (Ma et al.
2012;Peng et al. 2011), notably under severe human-forcing
Canadian Science Publishing
Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075 225
scenarios (Table 2). In southeastern NA, these changes will
be more moderate, but will nevertheless result in significant
PIMA growth reductions and mortality episodes (Girardin et
al. 2016b;Chaste et al. 2019). On the other hand, eastern
Québec and Labrador are predicted to become potential cli-
matic refugia for PIMA in the upcoming decades due to in-
creased temperatures and the absence of moisture limitation
(D’Orangeville et al. 2016). While some factors, such as in-
creased air and soil temperatures, may enhance PIMA pro-
ductivity over its range and partially compensate for growth
reductions due to moisture limitations, the increased impact
of late frosts, ESB, diseases, and parasites will introduce addi-
tional constraints on PIMA. These constraints should largely
result in abrupt growth declines and mortality episodes over
the majority of its range (Table 2). As all factors discussed in
this review are largely interrelated, they can also have many
cascading impacts, and make the response of PIMA to climate
change much more complex. For example, warmer soils may
lead to earlier budburst and thus increase frost damage (Li
et al. 2015) or enhance the impact of pathogens (Kliejunas
et al. 2009). Increasing ESB outbreaks could enhance the
impact of diseases and pathogens (Hudak and Singh 1970).
Furthermore, diseases and pathogens may reciprocally in-
crease the severity of ESB epidemics, but this mechanism has
been poorly investigated in NA boreal forests. Disease dam-
age may also be enhanced due to a decrease in soil insu-
lation caused by a thinner and shorter snow cover (Battles
et al. 2006). As the majority of these factors seem to rein-
force a negative eect or, at the very least, reduce a posi-
tive eect on growth and mortality rates, collectively they
should bring an additional constraint on PIMA growth with
climate change. Recent continental-scale modelling studies
confirm these assumptions, with a very large reduction in
suitable habitats for PIMA by the year 2100, particularly un-
der the RCP8.5 scenario (Prasad et al. 2020). A meta-analysis
conducted on all the species ranges could, however, help
to more precisely assess the future impact of all the fac-
tors discussed on PIMA growth and mortality rate in this re-
view.
Although this review focuses on the impacts of climate
change at the scale of the individual tree, we must men-
tion that climate-induced PIMA growth declines and mortal-
ity episodes will likely combine with ecosystem-scale distur-
bances, particularly fire, and reduce the potential produc-
tivity of NA boreal forests. In this sense, western and cen-
tral NA should again be the most impacted, with a large in-
crease in area burned (Flannigan et al. 2005), fire occurrence
(Wotton et al. 2010), and fire season length (Flannigan and
Wotton 2001) that may exceed the PIMA resilience threshold
and induce a shift away from conifer to mixed and deciduous
stands mainly dominated by aspen and birch (Baltzer et al.
2021). In eastern Canada, shorter fire return intervals should
result in a dierent postfire pattern and promote the tran-
sition from closed-crown to open-canopy forests (Girard et
al. 2008), making PIMA less likely to be aected by growth
reduction due to lower intraspecific competition (Ameray
et al. 2023) and less severe ESB outbreaks (Régnière et al.
2012). The exceedance of resilience thresholds by PIMA may
also be exacerbated in areas under active forest manage-
ment, where logging followed by fire may also increase the
regeneration failure of PIMA (Splawinski et al. 2019). This
species currently provides over 35% of the merchantable
wood volume for the Canadian provinces of Québec and On-
tario (Chagnon et al. 2022). Thus, human-related stressors,
such as forestry, are considered major disturbances in these
regions (Burton 2003) and are expected to have a signifi-
cant impact on the resilience of PIMA forests. For example,
clearcutting practices with short rotations may increase re-
generation failures, especially in regions marked by short
fire cycles or insect outbreaks, thereby exacerbating the loss
of PIMA resilience to disturbances (Splawinski et al. 2019).
More eective fire suppression, initiated in the 1970s (Lefort
et al. 2003), may at the same time mitigate increasing fire
frequency, but also enhance ericaceous competition with
PIMA seedlings (Mallik and Bloom 2005). Finally, as global-
isation and global trade accelerate, North American forests
are increasingly vulnerable to a growing number of biolog-
ical invasions. Among them, animal and plant species, as
well as diseases, also have the potential to directly aect
PIMA tree growth and mortality rates or work with previously
described biotic and abiotic factors, as well as forest man-
agement (Dukes et al. 2009). While North American boreal
forests currently face fewer biological invasions than south-
ern regions, Weltzin et al. (2003) suggest that climate change
could also intensify invasive species issues in the upcoming
decades.
Taken as a whole, our synthesis largely supports the idea
that future climate change will deeply and lastingly disrupt
the growth and mortality rate of PIMA, depending on its spa-
tial distribution. Although the entire PIMA range will be af-
fected, the predicted large decrease in water availability in
central and western Canada is expected to lead to signifi-
cant growth reductions and mortality events. While some
similar drought conditions may occur in its southern range
(D’Orangeville et al. 2018), the majority of eastern Canada is
expected to undergo a smaller water deficit resulting from
a lower precipitation–evapotranspiration ratio (Price et al.
2013;D’Orangeville et al. 2016), and thus be less impacted
than their central and western counterparts. However, there
will likely be latitudinal and longitudinal dierences in how
the Canadian provinces of Quebec and Newfoundland and
Labrador will be impacted. While southeastern NA should
be marked by the increased impacts of water stress, spring
frosts, and more frequent fire events and pathogen out-
breaks, the more northern regions should be spared. At the
same time, higher precipitation rates in the easternmost re-
gions of eastern Canada (particularly the Labrador Penin-
sula and the Côte-Nord region of Québec) are projected to
mitigate the increase in temperature, and therefore the re-
gion should remain favourable to the development of PIMA.
Thus, our synthesis supports the hypothesis proposed by
D’Orangeville et al. (2016) that the less moisture-limited re-
gions of northeastern NA may serve as potential PIMA refu-
gia in the upcoming decades. However, these refugia may be
temporary if future climate change becomes too severe, as
the conditions in northeastern Canada could become com-
parable to their western counterparts (D’Orangeville et al.
2018).
Canadian Science Publishing
226 Environ. Rev. 32: 214–230 (2024) | dx.doi.org/10.1139/er-2023-0075
Challenges for sustainable forest
management
Forestry is currently one of the most important economic
sectors in Canada, representing 1.7% of the national GDP in
2021 (i.e., $39.2 billion) (Government of Canada 2023), among
which PIMA represents one of the main economic resources
(Natural Resources Canada 2022b). The predicted major trans-
formations of the North American boreal forests, as described
in this paper, are unprecedented in human history and pose
major challenges to policymakers and forest managers from
ecological, climatic, and economic perspectives. Sustainable
forest management practices will likely need to adapt to meet
the increasing pressure of predicted climate change on bo-
real forests over the next decades, particularly in the south-
ernmost forest regions. Thus, proactive forest management
strategies should be implemented to mitigate growth reduc-
tions and mortality episodes of PIMA while reducing the risk
of fire exposure (Splawinski et al. 2019). As emphasised by
our synthesis, these changes in silvicultural practises must
be region-specific, since the impacts of climate change will
be far more significant in the central and western ranges of
PIMA as compared to their eastern counterparts. Mixed plan-
tations of PIMA and jack pine and an increase in seed tree
retention have been proposed as an interesting strategy to
maintain economically viable forest productivity (Cyr et al.
2022), but are only profitable in areas where the mean fire re-
turn interval is over 300 years (Splawinski et al. 2019), which
will not be the case over the majority of the PIMA distribution
according to recent models (Wotton et al. 2010). Thus, in ar-
eas where it is possible, the promotion of broadleaf species in
mixed stands is known to induce negative feedback and miti-
gate fire risk (Astrup et al. 2018), while increasing the produc-
tivity of PIMA during periods of water limitation (Augustin et
al. 2022;Ameray et al. 2023). However, Boulanger and Pascual
Puigdevall (2021) emphasise that since conifers are typically
preferred over broadleaves for wood supply, changes in for-
est composition will significantly impact the forest industry,
necessitating adaptation.
This review provides a good baseline to identify the chal-
lenges inherent in maintaining the resilience of PIMA and
adjusting sustainable forest management practices in NA bo-
real forests. It highlights the importance of both retrospec-
tive and current studies to identify the vulnerabilities of PIMA
to climate change and demonstrates that the eects of cli-
mate change will be significantly more negative for PIMA
growth and mortality rate under the RCP8.5 scenario than
lower anthropogenic forcing scenarios, particularly in west-
ern and central NA. Therefore, it underscores the increasing
importance of mitigating anthropogenic climate influences
to achieve temperatures lower than the predicted levels by
the end of the century. This action is essential to minimise the
impact of climate change on black spruce and, more broadly,
on NA ecosystems.
Acknowledgements
We thank the two anonymous reviewers for improving the
first version of the manuscript through their useful com-
ments and suggestions. We would also like to thank Paul
Jasinski for proof-reading this article.
Article information
History dates
Received:18July2023
Accepted: 23 January 2024
Accepted manuscript online: 2 February 2024
Version of record online: 13 May 2024
Copyright
© 2024 Authors Lesven, Druguet Dayras, Cazabonne, Gillet,
Rius, Bergeron, and © His Majesty the King in Right of
Canada, as represented by t