Thèse Modélisation des Effets du Réchauffement Climatique sur la Dormance des Graines d'Arbres Comparaison de Stratégies Physiologiques de Dormance Contrastées H/F - 33

Établissement : Université de Bordeaux
École doctorale : Sciences et environnements
Laboratoire de recherche : BIOGECO - BIOdiversité, Gènes & Communautés
Direction de la thèse : Marta BENITO GARZóN ORCID 000000023436123X
Début de la thèse : 2026-10-01
Date limite de candidature : 2026-05-27T23:59:59

Le changement climatique modifie les écosystèmes forestiers en perturbant les cycles biologiques, affectant ainsi la survie et la régénération des populations d'arbres. Alors que de nombreuses recherches se sont concentrées sur les arbres adultes, les premiers stades de développement-en particulier la germination et l'établissement des semis-sont plus sensibles aux changements environnementaux et jouent un rôle crucial dans la persistance des forêts. Ce projet de thèse examine comment les traits de dormance et de germination s'adaptent au changement climatique à travers de larges gradients environnementaux, en se concentrant sur deux espèces d'arbres aux types de dormance contrastés : le pin noir (Pinus nigra R. Legay) avec une dormance physiologique superficielle et le tilleul à petites feuilles (Tilia cordata Mill) avec une dormance physiologique et physique profonde.
L'étude vise à (1) quantifier les exigences de levée de dormance parmi les populations réparties sur l'ensemble de l'aire de distribution, (2) explorer les relations entre les besoins de levée de dormance, la germination et le développement post-germination, et (3) évaluer l'intégration des traits des premiers stades et leur plasticité en réponse au réchauffement climatique. Ces objectifs permettront de déterminer si les arbres peuvent ajuster le moment de leur germination et leur croissance aux conditions climatiques changeantes.
Des expériences de levée de dormance et un suivi de la germination seront réalisés dans des conditions contrôlées, simulant divers scénarios climatiques. Les traits post-germination-incluaient phénologie foliaire, contenu en pigments, allocation de biomasse et survie-seront mesurés sous différents traitements de température. La dynamique de germination sera analysée à l'aide de modèles de risque proportionnel de Cox, tandis que l'intégration phénotypique et la plasticité des traits seront évaluées par modélisation d'équations structurelles et réseaux de corrélation des traits.
Cette recherche fournira des informations essentielles sur la capacité d'adaptation des populations d'arbres, contribuant ainsi aux stratégies de conservation telles que le flux génétique assisté et améliorant les prévisions de régénération des forêts face au changement climatique.

Climate change is disrupting biological cycles, altering ecosystems and changing forests communities at an unprecedented pace. Tree populations can survive under climate change by moving towards more favorable conditions (Chen et al., 2011; Sunday et al., 2011), or persisting in-situ by evolutionary processes, including genetic adaptation and phenotypic plasticity (Valladares et al., 2014). These processes can be quantified from common gardens, where trees from different origins are planted all across the species ranges (Leites & Benito Garzón, 2023). However, current knowledge on tree populations' responses to climate change at large geographical scales is mostly based on adult trees, even though early developmental stages are far more sensitive to climate than adult ones and tightly linked to fitness (Verdú & Traveset, 2005). Therefore, new knowledge on the sensitivity to climate of the early phases of trees is urgently needed to assure forests persistence (Vicente & Benito Garzón, 2024; Figure 1).
Early tree stages encompass the very first processes of germination development, spanning dormancy breaking (for orthodox seeds, i.e. those seeds tolerating desiccation to moisture contents below 5%), germination rates and timing, growth, survival and phenology after germination. Most temperate trees present orthodox seeds allowing them to persist in bank soils and to be stored for long periods (Rosbakh et al., 2023). Among them, many present physiological dormancy, i.e. the physiological response inducing germination is controlled by environmental cues (Baskin & Baskin, 2001; Finch-Savage & Leubner-Metzger, 2006), therefore making them sensitive to any environmental change. Physiological dormancy is an adaptive mechanism that protect seedlings from frost damage during harsh conditions, and hence it has been widely developed in trees inhabiting cold and temperate regions with harsh winters. As adaptive trait, physiological dormancy can change among mother trees, populations and species (Andersson & Milberg, 1998; Kaye et al., 2018; Monemizadeh et al., 2021). Maladaptation to new climatic conditions can be expected in some cases, provoking lack of germination if cold conditions are not fulfilled. However, seeds may also adjust to new environmental conditions by physiological tolerance, i.e., phenotypic plasticity (Figure 2).
Indeed, germination-traits hold an important amount of intra-specific variation due to genetic adaptation and plastic adjustments to climatic clines across species' distributions (Cochrane et al., 2015; Donohue, Rubio De Casas, et al., 2010). As such, latitudinal variations in germination rates and speed across populations are common in orthodox (e.g. Accer saccharum, (McCarragher et al., 2011; Solarik et al., 2016); Betula pendula (García-Nogales et al., 2016; Solé-Medina et al., 2020). ) and recalcitrant seeds (e.g. Quercus ilex (García-Nogales et al., 2016; Solé-Medina et al., 2020), Quercus suber (Benito Garzón et al., 2024; Carme et al., 2026)). In spite of these examples, our knowledge on the cues affecting dormancy and germination in trees is very limited (Vicente & Benito Garzón, 2024). Furthermore, species and populations with orthodox seeds present a degree of dormancy spanning from shallow to deep physiological dormancy that differs strongly among populations and species, which suggest that dormancy is a an adaptive trait that adjust to changes in the environment (Poughon et al., 2025). However, to what extent their adjustments to warmer climates will allow them to first germinate and then survive under warmer conditions is still unknown.
In addition to germination rates, changes in germination timing will strongly disrupt species ecology. For instance, the timing of germination of dormant seeds requiring chilling and forcing temperatures to break dormancy before germinate is very disturbed by changes in temperatures (Gremer et al., 2020; Poughon et al., 2025). Both germination rates and timing have a high ecological relevance: germination rates directly influence the regeneration and the contribution of the seeds to the next generation (Baskin & Baskin, 2001; Donohue et al., 2010), whereas the germination timing determines the environment where the plant will develop, and hence its early-stage fitness (Gremer et al., 2020).
Furthermore, germination represents the first relationship of the plant with the environment, and it can have a cascade effect on other early-stage fitness-related and functional traits (Donohue, Rubio de Casas, et al., 2010; Gremer et al., 2020). For instance, the seed reservoir is mostly used by the cotyledons and hence bigger seeds tend to show faster shoot elongation and first leaf emergence, that can influence the overall seedling growth (Donohue, Rubio de Casas, et al., 2010; Vicente et al., 2025). Likewise, leaf protective pigments as flavonols and anthocyanins can confer protection to stressful environmental conditions as drought and cold temperatures (Chalker-Scott, 1999), in detrimental of other leaf pigments as chlorophyll. Indeed, trait co-variation modulates fitness across species ranges (Laughlin et al., 2020). The complexity of trait co-variation has led to the study of phenotypic integration, defined as the pattern of functional, developmental and/or genetic correlation among traits (Pigliucci, 2003). Phenotypic integration is a useful concept to understand how traits and their plasticity (i.e. individuals or populations adjustments to environmental changes) vary under large climatic gradients.
The main goal of this PhD is to understand the integration of early phenotypes (spanning those related with seeds, dormancy breaking, germination and postgermination) across large environmental gradients for two species with dormant seeds representing the extremes in the orthodox-seed gradient in Europe: Black pine (Pinus nigra R. Legay.; shallow physiological dormancy) and small leaf linden (Tilia cordata Mill.; deep physiological and physical dormancy). The expected results will bring us light on the adaptive and environmental adjustments of early stages traits along the dormancy tree spectrum to warming temperature, informing assisted gene flow programs and contributing to understand forest regeneration in a warming world.

1/ Previous analysis on dormancy breaking and germination showed that required time and temperature conditions for breaking seed dormancy vary within species, likely due to local adaptation and adjustments to climate along temperature gradients (Hudson et al., 2015; Poughon et al., 2025). The potential differences in dormancy breaking among populations are key to explore the assisted gene flow in the context of climate change and remains largely unexplored. Therefore, the first objective is to quantify dormancy requirements differences across range-wide populations of the two studied species.
2/ Understanding the relationship between the dormancy breaking needs, germination and seedling performance is essential to predict forest regeneration and management. We expect that warmer populations can germinate after shorter stratification (or without for shallow dormant seeds) than colder populations (i.e. they have adapted to warmer winters and hence are able to germinate under such circumstances). Those warmer populations may however produce seeds with less vigor and require longer times to germinate than stratified seeds, decreasing hence the overall fitness of the seedlings (Postma & Ågren, 2022). The second objective is to identify the relationship between the degree of dormancy and seedling fitness in across the dormancy gradient of the species (shallow versus deep dormancy).
3/ To predict the climatic niche of trees is essential to track the niches from germination and regeneration and contrast them with those of adult trees. We expect that regeneration niches are more sensitive to climate and hence more vulnerable to climate change than those of adult traits (Benito Garzón et al., 2024; Vicente & Benito Garzón, 2024). The third objective is to generation of germination and early stages climatic regeneration niches and compare to estimation of climatic niches based on adult sensitivity to climate.

1/ Species choice and seed collection. We selected two species with contrasting physiological dormancy: Pinus nigra with shallow dormancy and Tilia cordata with deep dormancy. Seed collection from 7 (P.nigra) and 10 (T. cordata) populations was performed in the framework of the EU-funded project OptFORESTS. Seeds were stored at -5°C at BIOGECO. Pinus nigra experiment started in January 2026 and the data will be ready to use at the arrival of the PhD student.
2/ Lab work. Laboratory work includes dormancy breaking experiments, germination monitoring and postgermination traits measurements under controlled conditions for Tilia cordata. We will break the seed dormancy at humid and cold conditions from 3 to 8 weeks to simulate different scenarios of winter cold. Afterwards, we will germinate the seeds in growing chambers in nursery trays in a mixed substrate of peat and sand. The nursery trays will be distributed across three climatic chambers located at INRAE BIOGECO-University of Bordeaux (Snijder LABS, micro clima-series) set at temperatures of 15, 20, and 25 °C, maintaining a constant 75% relative humidity, light intensity at 100 µmol m2 s1, and a photoperiod of 14/10 light/dark hours, emulating the conditions required for germination during springtime. During the dark phases, the temperature in all chambers was reduced by 7 °C. Seeds will be randomly positioned within each chamber. We will monitor germination every other day during 2 months, first leaf emergence, and height. Height, aboveground and belowground biomass, pigments content and survival will be monitored again one month after first leaf emergence. The same protocol has been used in the Pinus nigra experiments performed in January-June 2026.
3/ Statistical analysis and modeling. We will calibrate Cox's proportional hazard models of germination dynamics for each species using the daily germination data under different stratification scenarios (dormancy breaking) (Benito Garzón et al., 2024; Carme et al., 2026; Poughon et al., 2025). To analyze the early phenotypes we will use phenotypic integration techniques as structural equation models (Benito Garzón, 2021; Santini et al., 2019; Vicente et al., 2025) and trait correlation networks (Benavides et al., 2021). Climatic niche models will be performed with TraitSDM (Benito Garzón et al., 2019, 2024).