Climate change is increasing the intensity and variability of temperature and precipitation regimes throughout Canada (Zhang et al., 2019). As a result, forest model projections predict significant decreases in the abundance of cold-adapted boreal species such as balsam fir [Abies balsamea (L.) Mill] in the majority of their southern range from increasingly unfavorable growing conditions due to rising heat and drought (Boulanger et al., 2017; Taylor et al., 2017), while in the north, these species are expected to have generally large growth increases (D’Orangeville et al., 2018a; Wang et al., in press). The large potential ecological and economic impacts of such changes within Canada’s forests (Rodenhouse et al., 2008; Ochuodho et al., 2012; Rustad et al., 2012) require a better understanding of the adaptative capacity of the region’s most abundant tree species. Current observational studies of species responses to climate change have been asynchronous with some model predictions (Zhu et al., 2012; Fei et al., 2017), an issue that highlights the current lack of empirical data regarding age-specific and both inter- and intra-species responses to climate change across environmental gradients (Hijmans and Graham, 2006; Garzón et al., 2011; Aubin et al., 2016; Urban et al., 2016).

Low radial growth of balsam fir is correlated with low humidity (Duchesne and Houle, 2011), extreme low or high temperatures, and low soil moisture (Goldblum and Rigg, 2005; D’Orangeville et al., 2018a). However, seedlings differ from mature trees in their response to climate (Fisichelli et al., 2012, 2014; Gray and Brady, 2016) due to their small size and position in the understory (McDowell et al., 2008; Rollinson et al., 2021), and are at a higher risk for hydraulic failure and mortality from extreme climate events due to limited root volume and photosynthetic area (McDowell et al., 2008). Young balsam fir are particularly reliant on consistent soil moisture levels, likely partially due to the species’ proclivity for shallow rooting (Burns and Honkala, 1990; Taylor et al., 2020). When exposed to short-term moisture depletions, balsam fir seedlings show tolerance and plasticity in xylem growth and apical growth (Rossi et al., 2009), while long-term droughts decrease metabolic functions that limit growth in the current and following years (D’Orangeville et al., 2013; Vaughn et al., 2021).

While mean summer precipitation rates are projected to increase throughout the majority of balsam fir’s range (Zhang et al., 2019), higher frequencies of droughts and heatwaves will increase the prevalence of periods of low soil moisture, resulting in moisture stress and growth declines (Goldblum and Rigg, 2005; Way and Sage, 2008; Rossi et al., 2009; McDowell et al., 2011). Increases in atmospheric vapor pressure deficit (VPD) with warming and drying (Yuan et al., 2019) are expected to modify tree stomatal conductance and evapotranspiration rates (Novick et al., 2016; Grossiord et al., 2020). Limited soil moisture and a high transpirative demand due to warming can provoke stomatal closure in trees as a defense mechanism to mitigate water loss, therefore preventing cavitation by maintaining hydraulic pressure (Domec et al., 2009; Choat et al., 2018). In doing so, trees limit their rates of photosynthesis (Grossiord et al., 2020), a factor also directly inhibited by high temperature (McMurtrie and Wang, 1993; Lin et al., 2012). In turn, this reduces their capacity to produce valuable carbohydrate stores that may be used to mitigate cell death, cavitation, and hydraulic failure (Hacke et al., 2001; McDowell et al., 2011; Klein et al., 2018; Sapes et al., 2019). Long-term, persistent droughts eventually break the hydraulic tension in trees regardless of drought-resistance mechanisms, causing cavitation, which reduces the ability to transport water and solutes, leading to mortality (Allen et al., 2010; Choat et al., 2018).

All trees have a capacity to adapt to climate change, but the likelihood of a species to persist by both resisting and recovering from these climatic stressors is largely determined by its level and type of phenotypic plasticity and genetic diversity (Aubin et al., 2016). For instance, with limited moisture, some species lengthen their roots to access deeper moisture reserves, thereby increasing their root:shoot ratio (RSR), (Janiak et al., 2016). Trees can also adapt by increasing their photosynthesis to transpiration ratio, or water use efficiency (WUE), to reduce risk of carbon starvation and water loss (Osakabe et al., 2014), and refill embolized xylem to improve water conductance post-drought (Klein et al., 2018). Genetic adaptations resulting from local forcing events create adaptative differences between populations of species (Ghalambor et al., 2007), ensuring higher chances of survival under stress in certain populations (Moran et al., 2017). Research into the genetic variance amongst balsam fir populations, or “provenances,” has demonstrated high phenotypic variability within mature and sapling stage trees (Fryer and Ledig, 1972; Lowe et al., 1977; Carter, 1996; Akalusi and Bourque, 2021). In mature balsam fir, WUE was significantly different amongst provenances, with positive correlations between WUE and diameter at breast height (cm) (Akalusi and Bourque, 2021). A balsam fir provenance trial along an altitudinal gradient demonstrated sizeable variance of phenotypic plasticity among provenances through the adjustment of photosynthetic temperature optimums to better suit their origin climate, with a 2.4°C decrease in optimum temperature per 305 m increase in altitude (Fryer and Ledig, 1972). Incorporating these provenance-specific adaptations to climate into stand distribution models has been shown to greatly reduce the projected declines of climate change on a species’ abundance (Garzón et al., 2011). Therefore, empirically evaluating the effects of balsam fir seedling genetic variance and phenotypic plasticity on performance and survival under a range of environmental conditions is warranted to better understand the role of climate change in the boreal and temperate forests of North America.

We conducted a controlled seedling provenance trial within 12 climate-controlled greenhouse phytotrons with the objective of (i) determining the response of balsam fir growth and leaf-level gas exchanges to wide gradients of heat and drought and (ii) assessing the variation of these responses between provenances. To achieve these objectives, we evaluated the rate of physiological change of each provenance when exposed to five levels of drought intensity, and a 12-level temperature gradient, ranging from an average temperature of 13.9–30.9°C. To determine how seedling growth varied among treatments, we measured seedling biomass, RSRs, and CO₂ assimilation rates. We hypothesized that (i) rates of balsam fir seedling growth and photosynthesis will increase in response to moderate temperature increases but significant declines are expected under extreme warming conditions (i.e., a nonlinear response); (ii) balsam fir provenances from southern locations will experience less significant growth declines at higher temperatures when compared to northern provenances; and (iii) drought will limit the positive effects of moderate temperature increases, exacerbate heat-related growth declines, and increase seedling mortality.

Our experiment revealed considerable phenotypic plasticity in photosynthesis and RSR among provenances, which seemed to moderate balsam fir seedling stress under the warming and drought treatments. Although we cannot exclude that provenances from drier, western parts of the species range may behave differently than the provenances studied here, the results suggest that regardless of adaptations to local climate regimes, the physiology of the balsam fir seedlings studied here is capable of adapting to a warming environment resulting from climate change, as evidenced by the uniform growth declines among all provenances observed only in temperature conditions well beyond RCP 8.5 forcing scenarios within the warmest regions of the species biogeographic range.

Optimizing resource acquisition in limiting environments is essential for survival in seedlings due to their limited size. The observed RSR increases alongside temperature indicate an important potential ability of balsam fir seedlings to allocate resources to below-ground structures to regulate greater evapotranspiration, a strategy regularly employed by plants in moisture-limiting conditions (Lopez-Iglesias et al., 2014; Aubin et al., 2016; Janiak et al., 2016; Moran et al., 2017; Ledo et al., 2018). The plastic response of seedling photosynthetic levels to increasing temperature may enable seedlings a greater capacity to mitigate heat stress in needles (Haider et al., 2021), maintain hydraulic transport integrity through the shifting of xylem anatomy (D’Orangeville et al., 2013) or through the refilling of embolized cells (Klein et al., 2018), thereby sustaining crucial carbohydrate distribution throughout the system (McDowell, 2011; Kono et al., 2019). Conversely, high rates of photosynthesis may exacerbate plant moisture loss through stomatal conductance, a risky but potentially beneficial strategy in hotter climates (McDowell et al., 2008; Sade et al., 2012; Skelton et al., 2015). In hot growing conditions, this risky behavior may have been amplified by a complete lack of water-use efficiency (WUE) acclimation to any intensity of drought (see Supplementary material). Although the drought treatment had no impact on seedling performance, the late application may impact growth in the following year as the majority of the current-year growth had already completed (Goldblum and Rigg, 2005; D’Orangeville et al., 2018b; Kannenberg et al., 2019). Regardless, the observed positive effects from moderate temperature increases supports our first hypothesis, while the lack of effect from the drought treatment fails to support our third hypothesis.

The emergence of genetic variation can result from the selective pressures of heterogeneity in growing conditions within a species geographic range, and from geographic isolation (Via and Lande, 1985; Lande, 2009). Advantageous phenotypic plasticity resulting from this genetic variation can potentially improve the persistence of a population in uncertain conditions (Donohue et al., 2000; Chevin and Lande, 2010; Nicotra et al., 2010), potentially promoting further population divergence (Via and Lande, 1985; Kelly, 2019). Our results indicate a low to moderate level of variation in phenotypic plasticity between provenances, here represented as RSR and photosynthesis. As hypothesized, the southern provenances exhibited greater performance in hot conditions when compared to the northern provenances; however, these divergences had limited impacts on growth and needle damage rates overall. This low variation between provenances may relate to the theorized ecological cost of adaptive phenotypic plasticity, where adaptations to a specific environment could limit a species plastic response to further change or cause unintended consequences within plant responses to other stimuli (DeWitt et al., 1998; Van Kleunen and Fischer, 2005; Ghalambor et al., 2015). Alternatively, considering that the magnitude and type of plasticity can change alongside tree development (Bouvet et al., 2005), advantageous phenotypic plasticity may have a more noticeable influence on provenance growth variation within older trees that may be more prone to moisture stress due to a greater exposure to climatic extremes (Tyree and Ewers, 1991; Domec et al., 2008; McDowell and Allen, 2015; McGregor et al., 2021; Rollinson et al., 2021).

Model projections of balsam fir’s future range distribution (Iverson et al., 2008; Hassan and Bourque, 2009), abundance and productivity (Boulanger et al., 2017; Taylor et al., 2017) under varying climate change scenarios fail to consider age-specific provenance growth responses to climate, which contributes to uncertainty in model projections. Considering responses to climate found in previous studies (Carter, 1996; Akalusi and Bourque, 2021) and our own findings, it is important to further highlight the two factors in modeling approaches.

Although our experiment was conducted in a controlled greenhouse setting, stressful factors found only in natural growing conditions such as competition (Rollinson et al., 2016), pests, and pathogens (Allen et al., 2010; McDowell et al., 2011) may limit the observed positive growth response to temperature. Nonetheless, our research indicates that the southward extent of balsam fir is not directly limited by temperature, as seedling growth increases well beyond the current average temperature of the species’ southern range boundary. Recent research has also reported similar temperature-related growth benefits in young balsam fir in natural conditions (Collier et al., 2022), and no effect of winter warming on germination success (Vaughn and Taylor, 2022). Interestingly, the southward range of balsam fir saplings was observed expanding with climate warming, while mature fir migrated poleward (Boisvert-Marsh et al., 2014), suggesting that balsam fir may be more sensitive to climate stress in later stages of life. Though plagued with a risky hydraulic framework (Sperry and Tyree, 1990), shorter balsam fir have less distance to transport water therefore avoiding the risk of embolism and moisture stress that larger, mature trees face due to greater xylem tension, a higher exposure to drought conditions, and a greater moisture requirement (Sperry and Tyree, 1990; Tyree and Ewers, 1991; Domec et al., 2008; McDowell and Allen, 2015; Aubin et al., 2018; McGregor et al., 2021; Rollinson et al., 2021). The moisture dependence of balsam fir has been noted in previous studies (D’Orangeville et al., 2013, 2018a; Collier et al., 2022), and may be explained by the growth efficiency trade-off between tracheid embolism resistance and hydraulic conductivity; in non-moisture limiting conditions, a low conductivity may reduce competitiveness (Sperry and Tyree, 1990; Tyree and Ewers, 1991; Domec et al., 2008). Therefore, the generally hydric conditions throughout the range of balsam fir promote a high-risk hydraulic framework, fast growth, and a subsequent moisture dependency, thereby potentially reducing climate resilience and competitiveness in conditions with high hydrological variability.

All balsam fir provenances exhibit a consistent, striking ability to acclimate their physiological traits via phenotypic plasticity, enabling high growth in considerable heat. This suggests that temperature and drought may not be the limiting factors that moderate the establishment and growth of balsam fir seedlings at range boundaries, instead, climatic stress may be exacerbated with age and cumulative abiotic and biotic stressors, thereby influencing the reproductive success and competitive vigor of older individuals. Although the implications of our research are limited by the controlled nature of our experimental design and can only be generalized to the eastern, wetter part of the species range where the species is most dominant, these findings support the integration of age- and provenance-specific growth responses into future modeling attempts aiming to evaluate species distributions and abundance under climate change scenarios. To do so, more empirical data relating to these interactions needs to be collected in both controlled and natural environments. Furthermore, while our research aimed to evaluate how range wide temperature heterogeneity influenced the variance of genetics and phenotypic plasticity between provenances, future research should incorporate greater provenance diversity to increase our holistic understanding of balsam fir climate resilience. Future research should also be conducted to decipher how competition, legacy effects, provenance origin, and biotic stressors influence the success of balsam fir under differing climatic conditions, and at different life stages.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

JR performed the experiments, analyzed the data, and prepared the manuscript. All authors conceived the study design and contributed to the final version of the manuscript.