At the interface of ecophysiology and functional ecology, our research involves different disciplines (chemistry, geography, mathematics), tools (simulation and statistical modelling) and playground (tundra, taiga, temperate grasslands and forests, wetlands and drylands), but can be sumed up through four main contributions, presented below:

Contribution 1 – Refining strategies of photosynthesis within and across habitats

Understanding adaptation of photosynthesis to environmental constraints is still one of the most discussed topics in ecophysiology. In collaboration with international teams and using global plant trait datasets (TRY & Globamax) that I contributed to create (A31, A17), I worked on two theories of photosynthesis. Through the theory of the coordination of photosynthesis (A8), we demonstrated that the photosynthetic machinery is actively co-regulated within habitat to adapt radiation capture with CO2 capture. Within habitat and irrespective of the vegetation type, species displayed different photosynthetic rates but functioned with the same CO2/radiation capture ratio. Based on the optimization principle, we also showed that across habitats this CO2/radiation capture ratio can be predicted from the average growing conditions. Through the least-cost theory of photosynthesis (A13), we could predict the CO2 capture vs H2O expenditure strategy that best minimizes the entire plant respiratory cost according to environmental stress. These theories are powerful to improve Earth System Models and vegetation dynamic modelling, as they drastically decrease the number of parameters required for modelling photosynthesis. Thus, they help to better predict carbon, water and nitrogen dynamics at both community and global scales, and to detect unexpected covariations of photosynthetic traits across habitats. We notably showed that the disruption of such photosynthetic trait covariation was an early signal of desertification (A30).

Contribution 2 – The multidimensional niche of plant species

Plant functional traits determine the functional niche of species as they define how species both respond to and modify their biotic and abiotic environment. We first showed that the number of niche dimensions is higher than the usual two ones (leaf and wood economic spectrums) that define the classical ecological strategies contrasting species dominance across habitats. For instance, we demonstrated the existence of independent trade-offs, such as the ammonium vs nitrate uptake preference (A3), the early vs late leaf emergence, and the tissue elongation vs spreading development (A9). More importantly, we highlighted that such independent niche dimension is not related to differentiate species dominance across habitats (habitat-filtering process), but was rather to how species coexist within habitat (niche differentiation process). We further demonstrated that such niche dimensions associated with species coexistence are where rapid natural selection occurs (A28). All these results highlighted the need to better describe the niche of species. This work has been acknowledged by the peer community, even in animal ecology (e.g. to explain the assembling of stream-breeding anuran communities) and in cellular biology (e.g. to describe the coexistence of cancerous cells with healthy ones).

Contribution 3 – Towards a better understanding of the influence of plant communities on ecosystem functioning

Community traits have been viewed as cornerstones to link biological organisation scales. A community trait is usually calculated as a weighted-mean, which likely emphasizes the functional dominance of trait values. As normal distribution is more the exception than the rule in Ecology, we showed that community weighted-means do not necessarily describe a community properly (A18).

We showed that other properties of trait distribution held critical information to understand how plant community influences ecosystem functioning (A15). For a given trait, we showed that the functional dispersion, rarity and evenness of trait values vary independently, and are powerful predictors of leaf litter cycling (A34). Moreover, we investigated the distribution of multiple traits within communities and across wetlands (bogs, fens, humid grassland, marsh). It highlighted notably that trait patterns within and across wetlands are better detected using the community dispersion of trait values than the community means (A42).

Contribution 4 – Importance of soils in shaping functional strategies of plants and micro-organisms

Compared to climate, soils have received little attention from studies about the sources of variation in plant and micro-organism functional strategies. This is notably because climate is an important driver of soil formation. However, my team highlighted that soils, either independently or in conjunction with climate, influence significantly the functional strategies of plants and microbes related to resource use for photosynthesis, soil respiration and soil phosphorus cycling (A14, A21,A24, A38,A39). For instance, collaborating with an Australian network, we show that soil pH as a key property favoring nitrogen nutrition and photosynthetic rate (A33). As such, we revealed the how, i.e. the key properties of soils that influenced biological processes, and the where, i.e. the location where functional strategies are more influenced by soil properties than climate. We geospatialized our results, which allows generalization and utilization of our result at very large scale. As a result, other studies compared their results at local scales to our global available predictions.

Contribution 5 Feedback effects of Arctic vegetation on active layer and permafrost

Global change has drastically influenced environmental conditions of tundra ecosystems and the functioning of Artic plants. Within two FRQNT team projects (PI-2017, co-PI-2020), my colleagues and I focused on how Arctic plant communities respond to multiple indirect drivers of global change (c.f. nutrient availability, water availability, herbivory) and, in turn how plant communities modify tundra functioning. We first showed that indirect drivers could influence plant species and tundra functioning as much as global warming (A37). Second, traits related to vegetation cover and mineral nutrition were the most responsive traits to these environmental drivers (A35). Interestingly, we also revealed that these traits had a significant influence on soil carbon storage and permafrost thaw (A40). For instance, the accumulation of vegetation cover subjected to herbivory exclosure or higher nutrient availability led to the formation of an isolating layer at the top of the soil, which limited heat energy transfer to permafrost and decreasing the thawing front over the entire year. The influence of increasing erect shrub abundance on soil organic carbon storage represents another clear example of the feedback effect of vegetation change on permafrost soil. We found that the erect shrub Salix richardsonii led to organic matter accumulation in the soil only under low plant demand in nitrogen to the soil, thus highlighting strong vegetation feedbacks on carbon storage with nutrient cycling (A41).