Constellation, le dépôt institutionnel de l'Université du Québec à Chicoutimi

Résumé

In temperate and boreal ecosystems, trees shift between the phenological phases of growth and dormancy in order to avoid frost during the cold season. This cycle includes changes in frost hardiness, which is at a minimum during the growing season and reaches its maximum during dormancy. As climate change intensifies, warming temperatures and increased weather variability may create a mismatch between the phenology of locally adapted populations and their surrounding environmental conditions, leading to increased risk of frost damage. Field studies exploring the link between phenology and frost hardiness are still missing for many tree species, despite the fundamental importance of these traits for tree survival in cold regions. One particular source of uncertainty is the role of intraspecific differentiation, as different provenances may exhibit diverging responses in their phenology and frost hardiness, resulting in different risks of frost exposure. In this work, I used a combination of field observations and lab analyses to link phenology and frost hardiness in two key species of Eastern Canada’s forests, black spruce (Picea mariana Mill. BSP) and sugar maple (Acer saccharum Marsh.). For each species, different provenances were compared to account for intraspecific variation. The general objectives were i) assessing intraspecific variability in frost hardiness and phenology of budbreak and dormancy, and ii) linking these two traits to understand intraspecific variations in risk of frost damage. The study is articulated in three chapters, each focusing on a specific objective and hypothesis. Chapter 1 focused on the impacts of a natural late frost event occurring in a P. mariana common garden, analyzing frost damage through the lens of intraspecific differences in budbreak timings. Most of the provenances were in warmer conditions compared to their climates of origin, since the common garden was situated at the southern limit of the species’ range. The hypothesis was that under the same conditions, provenances from colder climates were more damaged by late frost because of an earlier budbreak, which exposed growing shoots to freezing temperatures. The results confirmed the hypothesis, with northern P. mariana provenances performing earlier budbreak consistently over several years of observations, resulting in higher damages during the late frost event of 2021. The study demonstrated how intraspecific differences in budbreak phenology can determine different risk of frost exposure within the same species, and highlighted how forest management can decrease the risk of late frost by selecting provenances with later spring phenology. Chapter 2 analysed frost hardiness in seedlings belonging to seven A. saccharum provenances growing in two sites near the northern limit of the species’ range, using repeated destructive sampling to measure frost hardiness during the winter season and compare its endogenous and environmental controls. The hypotheses were that 1) local weather influences frost hardiness, with seedlings in the colder site showing earlier and faster acclimation in autumn, higher maximum frost hardiness during winter, and later and slower deacclimation in spring, compared with the southern site; and 2) climatic conditions at the provenance origin influence frost hardiness, with seedlings from colder provenances showing earlier and faster acclimation in autumn, higher frost hardiness, and later, slower deacclimation in spring compared with warmer provenances. The results partially confirmed hypothesis 1, as the saplings in the warmer site showed earlier and faster deacclimation confirming the strong effect of warm temperatures on spring reactivation. Hypothesis 2 was rejected since no difference was found during acclimation in the autumn or maximum frost hardiness during winter, nor between provenances. LT50, i.e. the lethal temperature for 50% of the cells, varied between -4 °C in summer (July) and -68 °C in winter (February). All provenances attained LT50 of -55°C or lower, far below the common minimum winter temperatures at the northern limit of the species’ distribution. Chapter 3 focused on the role of freezing temperatures on chilling accumulation and endodormancy break. A. saccharum saplings belonging to 7 provenances and growing near the northern range limit of the species were exposed to either natural or artificial chilling conditions (4°C) starting at the end of autumn. Samples were transferred to forcing conditions v at regular intervals throughout the winter. I measured frost hardiness at the time of transfer and observed time to budbreak under forcing conditions. The predictions were that 1) endodormancy break would be easier to detect in artificial chilling treatments, where the confounding effect of frost hardiness would be limited; 2) a chilling model considering freezing temperatures would be more effective in a cold climate experiencing temperature below 0°C for several months; 3) samples with higher frost hardiness would take more time to perform budbreak, as they need more time to deacclimate. The results confirmed our hypotheses, highlighting how 4°C can both fulfill the chilling requirement and promote deacclimation, leading to ontogenetic development in the bud. Conversely, samples under natural chilling retained higher frost hardiness until late in the season, correlating with longer time to budbreak and a more difficult identification of an endodormancy break point. Provenance did not have a significant effect in the time to budbreak under forcing conditions. My results indicate that including freezing temperatures can improve chilling calculations in cold climates, where temperatures remain below 0°C during most of the winter. Moreover, measuring frost hardiness during chilling-forcing experiments can clarify its role in chilling accumulation and dormancy dynamics. The three chapters of this work added original data to the literature, providing insight into the interconnected traits of frost hardiness and phenology. The results show how tree species can have different degrees of plasticity and intraspecific differentiation. P. mariana transferred to warmer conditions had different timings of budbreak between provenances, a trend consistently observed over several years of observation. On the other hand, A. saccharum had higher plasticity, with provenances from the study area (the northern portion of the species’ range) showing similar responses to environmental conditions in both frost hardiness and budbreak. These intra-specific differences in plasticity have important implications for frost risk. Northern P. mariana provenances may be more exposed to late frost in the future because of their lower forcing requirements, but provenance selection may be an effective tool to reduce the risks, at least in commercial forestry. A. saccharum may already be able to survive winters north of its current range of distribution thanks to its high frost hardiness. Its increased plasticity in frost hardiness and budbreak means that provenance selection would have little effect on frost risk reduction. This work shows the importance of studying frost hardiness and dormancy regulation, especially under global warming. These traits are fundamental components of the life cycle of trees, and a more thorough comprehension is needed to improve predictions of climate change impacts and to develop adaptive forest management practices.