Project Topics
Get Complete Project Material File(s) Now! »
Species isolation and endemicity: consequences for adaptation
Detailed characteristics and species evolution in islands have been previously discussed in Weigelt et al. 2013; Warren et al. 2015; Patiño et al. 2017 (among others). Essential features that might condition plant adaptation to environmental changes are the limited habitats available to track optimal conditions within the island. If the landscape is heterogeneous enough, it might allow micro-scale tracking (Harter et al. 2015; Irl et al. 2015). However, this process might not be sufficient to cope with large or intense change (Chen et al. 2011; Moritz & Agudo 2013). For example, opportunities for an altitudinal shift by species might be limited as islands tend to have low elevations, except for the largest ones (Weigelt et al. 2013). Furthermore, compared to continental ecosystems, oceanic islands have relatively low species numbers and simpler community structures. When a change occurs, biotic interaction and resources are expected to favour ecosystem resilience which might be less the case in island ecosystems (Harter et al. 2015). When a change such as rapid climate change occurs, it might strongly affect native island flora. For example, in the Azores islands, more than 90% of native plant species are expected to lose s habitat, with for some of them the loss of all adequate habitats (Ferreira et al. 2016). Species already found at the upper part of islands are expected to be even more threatened in Faeroe Islands, among other locations (Fosaa et al. 2004; Harter et al. 2015 and references therein). According to this, the ecological integrity of island ecosystems is relatively sensitive. Islands harbour a much higher proportion of endemic species than continents, and island endemic species are especially affected due to their limited opportunities to escape unfavourable conditions (Harter et al. 2015). Endemic species result from local speciation or reduction of a formerly larger geographical area and might result from shorter or longer evolution related to taxon cycle with phases of expansion and extinction (Ricklefs & Bermingham 2002; Ricklefs 2011). Endemic species are strongly associated with particular habitats, usually extremely stable (Jansson 2003). In this regard, it is interesting to consider « phenotypic integration », i.e. the number and strength of trait correlations within an organism (Pigliucci 2003). A strong positive correlation was found between phenotypic integration and the degree of endemism of 14 species across an endemism gradient to Kerguelen Islands (Hermant et al. 2013). The question was whether high phenotypic integration found in the most endemic species might limit their phenotypic plasticity when facing changes (Hermant et al. 2013). It was suggested that strong correlations among traits might lower the ability of organisms to reach an optimal trait value in response to the change. Further studies targeting the intraspecific phenotypic variability of Kerguelen Ranunculus sp. and the Kerguelen Cabbage, Pringlea antiscorbutica, showed positive correlations between phenotypic integration and phenotypic variability (Labarrere 2017). Therefore, more ecological and evolutionary studies are needed to understand the fate of endemic species under climate change (Cartwright 2019), and this might be even more challenging for these plants in harsh environments.
Rapid climate change and consequences for plant species
Alpine, sub-polar and polar regions are subject to rapid climate changes, probably the fastest-warming on Earth (IPCC: Ranasinghe et al. 2021). These regions face a temperature increase, precipitation variation (quantity and regime), and modification of wind regime, among other factors (IPCC: Ranasinghe et al. 2021). For the end of the century, prediction for the whole sub-Antarctic is an increase in temperature and precipitation variations. However, the intensity of temperature increase is island-specific, from 1°C in Heard and Macquarie Islands up to 1.5°C in Marion Islands for example, when considering the greenhouse gases representative concentration pathway (RCP) 6.0, mainly due to the position of these islands in relation to the southern annular mode (Harter et al. 2015). This temperature increase may look small, but it is very large for such buffered climates with both low temperature ranges and low local averages. Moreover, the decrease in precipitation and the increase in evaporation should increase the salinity in the sub-Antarctic Islands (le Roux & McGeoch 2008). The intensity of the temperature, precipitation and wind changes will impact the possible response of the species to that change (Moritz & Agudo 2013). This is of special concern as alpine, sub-polar, and polar ecosystems often have low heat resilience (IPCC: Ranasinghe et al. 2021). Impact on long-lived species was highlighted with the occurrence of dieback in various harsh environments facing rapid climatic changes (Molau 1996; Frenot et al. 1997, Kleier & Rundel 2004; Bergstrom et al. 2015; Bjerke et al. 2017; Dickson et al. 2020, among others). Dieback is defined as the process of expansion of mortality where a particular portion of the cushion or the entire cushion fails to regenerate new foliage (Armesto et al. 1980; Whinam et al. 2014). Necrosis is the visible damage at a particular time and can be a symptom of the dieback process (Whinam et al. 2014). In the sub-Antarctic, the most documented case concerns Azorella macquariensis (Apiaceae) (Fig 4 A and B), a long-lived endemic cushion plant of Macquarie Island that shows rapid dieback all over the island due to a combination of abiotic factors and pathogens enhanced by climate change (Bergstrom et al. 2015; Dickson et al. 2020), (Fig 4 B). In this species, the mean dieback increased significantly by 30% between 2009 and 2011 (Fig 4 C). A rainfall exclusion experiment on a congeneric long-lived cushion plant, Azorella selago from Marion Island, resulted in necrosis after a year of this treatment (Le Roux et al. 2005). Therefore, these keystone species will very likely be vulnerable to climate change. However, studies addressing the adaptive capacity of these plants remain few.
Aims and hypotheses of the PhD
The PhD work takes place in the broad framework of assessing in-natura plant vulnerability and their adaptive capacity to climate change. Little research has concerned long-lived species, and even fewer studies have been done in sub-Antarctic Islands despite this region facing some of the most rapid climate changes on Earth. The Kerguelen Islands’ fellfields harbour a long-lived endemic cushion plant, Lyallia kerguelensis, which might have a low adaptive capacity as necrosis damage, possibly due to climatic stress, was recorded for a few decades. In this context, we raise the question: what is the adaptive capacity to climate change of the Kerguelen long-lived endemic cushion plant species, Lyallia kerguelensis,?
L. kerguelensis, by its long lifespan, cushion form, and endemism to isolated islands characterized by harsh environments subject to rapid climate change, appears as a good model. Its pool of variability (morphology, transcriptome, necrosis extent and soil rhizomicrobiome) across contrasted environments might provide insights into L. kerguelensis adaptation to harsh environments and its possible responses to changes. Plant morphology is one of the traits that respond the most to environments. Therefore, the pool of morphological variability of L. kerguelensis and its relation to specific environmental variables might be essential to identify the species phenotypic diversity and within it the possibility of a phenotype to be appropriate to the new condition (Chapter 1). At a thinner scale, it will be essential to find which genes or pathways are differentially expressed in L. kerguelensis in contrasted environments to decipher or reveal population biological and cellular responses possibly linked to local adaptation (chapter 2). When considering the extended phenotype, L. kerguelensis rhizomicrobiome assembly filtered from the surrounding microbiome might inform us on the selected generalist and specific microorganisms, apt to sustain plant vigour in a contrasted environment (chapter 3). Integrating temporality is important, for long-lived species such as L. kerguelensis, monitoring morphological trait dynamics within a number of years in contrasted environments might be essential to estimate their responses and the probability to cope with fast climate change (chapter 4). Finally, plant growth rate and age in various environments might influence adaptation outcomes as they underlie plant persistence and resilience to climate change (chapter 5). These findings will broaden our knowledge of the adaptive capacity of long-lived plants in isolated harsh environments facing rapid changes. It will allow to better assess their vulnerability to climate change, and it will be valuable for evolutionary and conservation research.
Allometry of traits and morphological variability
The surface area, perimeter, and height of the cushions were strongly positively correlated one to each other (Table 3). However, the cushion compactness was correlated only with the cushion height and the cushion shape was correlated to none of the other morphological traits. Mean cushion shape ratio (relation between short and long diameter) was 0.75 (n = 319, sd = 0.01) with low variation indicating that cushions approximate an ovoid cushion shape. Moreover, the ratios between height and short or long radius were also close to 1 (n = 279, mean ± sd; 0.85 ± 0.03 and 1.17 ± 0.04 respectively). For vigorous and non-vigorous cushions the relationships between the necrosis extent and the three morphological traits (cushion surface area, shape, and compactness) were not significant.
Relationships between cushion morphology and environment or geography
The variability of the cushion surface area across populations was well explained by the topography and habitat, slope aspect and soil texture models, explaining 75%, 54% and 19% of overall variation respectively (Table 4). In the first model, two conditions were significant, topography and wind exposure, without significant interaction. A post-hoc test showed that larger cushions were present mid-slope rather than on flat areas (Fig. 3a). A non-significant trend of larger surface area with a decrease of wind exposure was found. For the slope aspect model, the post hoc test did not identify any significant slope aspect influencing the cushion surface area. Finally, soil texture was positively related to the cushion surface area, with a larger percentage of coarse sand being correlated to a larger cushion surface area . Inter-population variability in cushion shape was explained (52% of variation) by the univariate model with slope aspect variable (Table 4), with a lower circular shape preferentially observed with south-east slope aspect compared to locations with south-west or north-east slope aspects or flat locations (Fig. 3b). Besides, variability in cushion compactness was significantly related to the surface texture model (18% of variation; Table 4), and was positively correlated to the proportion of bare soil (Pearson correlation, r = 0.478, n = 19, p= 0.039). To determine relationships between population morphology and geography, we applied Mantel tests separately for each morphological trait. For cushion surface area and shape the results of the Mantel tests were insignificant, respectively, r = -0.178, p = 0.933 and r = -0.022, p = 0.441. The results were significant only for the cushion compactness (r = 0.320, p = 0.011), where longer distances across populations were related to higher inter-population variability of cushion compactness but not of cushion shape or surface area.
Relationships between environment, geography and morphology
The surface area of cushions was related to topography and wind exposure, with larger cushions predominantly developing mid-slope rather than on flat areas, and a trend of larger cushions in less wind-exposed populations. In the fragmented landscape and harsh climate of Kerguelen, topography and wind are two major environmental components. The hilly topography of the island (Fig. 2) generates heterogeneity in rainfall, water retention and wind exposure (Aubert de la Rüe 1964; Wagstaff and Hennion 2007b). Mid-slope areas receive runoff water from higher altitude and at the time drain to lower altitude, unlike flatter areas, while wetlands usually develop at the base of slopes. The Kerguelen Islands, like other sub-Antarctic islands, are considered to have moist climates (Hennion et al. 2006b) and sub-Antarctic plants are known for their requirement of water for growth (Dorne and Bligny 1993; Hennion and Walton 1997). Thus, mid-slope may provide optimal water supply for the growth of L. kerguelensis. Furthermore, wind exposure induces mechanical stress and can accelerate transpiration of plants (Körner 2003; Haussmann et al. 2009; Gardiner et al. 2016). Plants from L. kerguelensis developing in less windy environments such as mid-slope may be less subject to extreme meteorological events like storms or severe frost. Finally, contents of coarse sand in the underlying soil was positively correlated to cushion surface area. Sandy soils are generally well draining, with low water storage capacity and a high usable fraction (Körner, 2003). This finding suggests that ease of water uptake favours larger cushions. Variation in cushion shape between populations was mainly explained by slope aspect. In Kerguelen, north and north-west slope aspects provide the greatest solar exposure but are also the most exposed to wind, and vegetation is more luxurious away from these slope aspects (south or east aspect) (Werth 1911; Aubert de la Rüe 1964). Our data indicated lower occurrence of circular cushions on the south-east aspect, less exposed to wind, suggesting that the circular shape might be associated with wind exposure in L. kerguelensis. This finding is in agreement with the recent air flow dynamics model on A. selago where cushions with more pronounced crescent shapes were in habitats with less airflow turbulence (Combrinck et al. 2020).
Variability in cushion compactness was partly explained by the proportion of bare soil and geographic distance between populations. Greater cushion compactness was correlated with a higher proportion of bare soil. This may indicate lower protection against wind than on rocky ground. Thus, higher compactness in cushion plants generally may indicate greater exposure to cold or desiccation stress. It is interesting to note that compactness in cushion plants generally co-varies with altitude, that may be correlated to temperature and humidity variation (He et al. 2014; Cranston et al. 2015; Zhao et al. 2018). However, L. kerguelensis did not show compactness variation across the altitudinal gradient examined here. In summary, cushion surface area, shape and compactness were all driven primarily by two major environmental variables: wind intensity and water availability.
Table of contents :
Plant adaptation: What does theory say?
The adaptive capacity concept
Extended phenotype and adaptive capacity
Species isolation and endemicity: consequences for adaptation
Life in harsh cold environments
Rapid climate change and consequences for plant species
The Kerguelen Islands environment
Geographical location
Abiotic and biotic characteristics
Climate and climate change
Lyallia kerguelensis (Montiaceae), a long-lived cushion plant species endemic from the Kerguelen Islands and affected by necrosis
Biology
Ecology
Ecophysiology
Aims and hypotheses of the PhD
Organisation of the thesis
References
PART I: Morphological variability and gene expression in Lyallia kerguelensis in contrasting environments
Chapter 1: Morphological variability of Lyallia kerguelensis in relation to environmental conditions and geography in the Kerguelen Islands: implications for cushion necrosis and climate change
Abstract
Introduction
Materials and methods
Results
Discussion
Conclusions
References
Online Resource
Chapter 2: Differential gene expression of L. kerguelensis in contrasted environments: implications for biotic and abiotic stresses
Introduction
Material and methods
Results
Discussion
Conclusion and perspectives
References
Supplementary material
Supplementary tables and figures
PART II: Soil microbial diversity and composition and its relation with rhizomicrobiome diversity and composition and plant vigour in contrasted environments
Chapter 3: Fellfields of the Kerguelen Islands harbour specific soil microbiome and rhizomicrobiomes of a long-lived endemic cushion plant facing necrosis
Abstract
Introduction
Material and methods
Results
Discussion
Conclusion
References
Supplementary data
PART III : Temporality of L. kerguelensis cushions’ morphological change and their possible longevity in relation with abiotic and climatic factors in contrasted environments
Chapter 4: Plant-climate relations and resilience to change: insights from a long-lived sub-Antarctic cushion plant (Lyallia kerguelensis)
Abstract
Introduction
Materials and methods
Results
Discussion
Conclusion
References
Supplementary data
Chapter 5: Growth and age variability in contrasted environments and architectural traits allometry in Lyallia kerguelensis Hook. f. (Montiaceae)
Abstract
Introduction
Material and Methods
Results
Discussion
Conclusion
References
Supplementary data
General discussion
Pool of variability of L. kerguelensis
Towards an estimation of the adaptive capacity of L. kerguelensis
Inter-population or region variability
Inter-individual variability
Long life-span and regeneration
A scenario under strong climate change
General conclusion
Perspectives of the thesis work
Towards a better understanding of the cushion entity
Shift of fellfield functioning with L. kerguelensis as a model plant
References
Annex 1: Hydric potential of Lyallia kerguelensis
Context
Materials and Methods
Results and Discussion
References
Annex 2: Germination experiments
Context
Material and methods
Results and discussion
Conclusion
References
Annex 3: Scientific outreach
PhD Publications
Other Publications
Communications
Scientific vulgarisation
Teaching experiences
Classes followed during the PhD
