Disentangling the relative importance of environmental conditions, dispersal and biotic interactions using empirical data require adequate methods that would jointly account for the effect of space, environment and the biotic context. The last few years have witnessed an upsurge of novel statistical developments that have the potential to elucidate this dilemma. These joint species distribution models (JSDMs, Warton et al., 2015; Ovaskainen et al., 2016) predict species distributions based on environmental and spatial variables (as in typical species distribution models), but they allow species to share information on their response to environmental or spatial variables and additionally consider the effect of all other co-occurring species. JSDMs are a specific kind of hierarchical models within a generalized linear model framework (Pollock et al., 2014). They allow to model multi-species-level data hierarchically at the species/functional group level and take into account potential dependencies between species. For these reasons, JSDM frameworks should be capable of generalizing and unifying approaches classically used in community ecology and conservation biology such as the analysis of biodiversity indices (e.g. species richness), multivariate analyses (e.g. canonical analysis) and species distribution models. The potential advantage to this approach is a quantification of how much species correlation is due to the environment and how much is due to variation not captured by the environment (e.g. residual correlation)(Ovaskainen et al. 2010; Pollock et al. 2014). However, this residual correlation might be due to ecological or evolutionary interactions (Pollock et al. 2015), dispersal limitation, or simply missing environmental variables.
This PhD will make use of these novel developments to get a better understanding of the drivers shaping the assembly of plant communities along environmental gradients in the European Alps. This will take place within the Cross-Disciplinary Project Trajectories (WP3) funded by the Univ. Grenoble Alpes, in which plant diversity trajectories will have to be modelled in function of climate and land use changes.
As a first step, the PhD will investigate whether JSDMs are able to capture positive and negative species interactions within a single trophic level focusing on plant communities. It will be done by analysing large-scale empirical datasets (plant community plots across the European Alps, Forest inventory plots in the French Alps) as well as fine resolution data (Orchamp data collected within the Trajectories project). To evaluate whether the correlation matrices capture real biotic interactions in forests we will confront them with estimates of biotic interactions based on neighbourhood modelling of growth and survival. We will test if species pairs with strong negative correlation are the ones with estimates of strong competitive interactions based on growth. The residuals correlation matrices will also be confronted to trait and phylogenetic distances between plant species. Under strong limiting similarity mechanisms (e.g. competition), we expect that negative residuals from JSDM correlate with small trait distances between species (i.e. similar species compete more strongly than different species). Species with similar traits will be expected to co-occur less than expected under similar environmental conditions. The comparison of analyses across scales and response variables will allow identifying at which scales community assembly processes leave the strongest signals in distribution patterns. The PhD will then investigate how the direct integration of phylogenetic relatedness and functional traits in the JSDM can help retrieving community assembly processes and provide more reliable predictions.
In a second step, we will investigate how JSDM can capture biotic interactions occurring across very different types of vegetation and across trophic levels (in comparison to the analyses of step 1 that focus on one single trophic level and vegetation type). For this step, we will focus on the Orchamp data collected within the Trajectories project for which grass and tree communities are monitored, but also soil communities through innovative environmental DNA approaches. Thus, these data allow investigating biotic interactions across grasses, trees, and the soil communities based on traits and phylogenetic relationships to gain a better understanding of community assembly but potentially also making more reliable predictions.
In a final step, the PhD candidate will use these integrated models to predict the future state of plant communities in the Alps in function of both climate and land use change. He/she will investigate how the uncertainty in the predictions decreases with the inclusion of additional information such as traits or
phylogenetic relatedness and other trophic levels. Current and future state of plant communities will then be translated into changes in ecosystem functioning and ecosystem services.
The PhD candidate will be supervised by W. Thuiller at LECA Grenoble and B. Reineking at IRSTEA Grenoble. Amongst the closest collaborators, T. Münkemüller, L. Pollock and M. Talluto at LECA and G. Kunstler at IRSTEA will actively contribute to the PhD mentoring. Nick. Zimmermann from WSL Zurich will also contribute.
The candidates must have a Master’s degree (or be about to earn one) or have a University degree equivalent to a European Master’s (5-year duration), in a relevant area (ecology, ecological statistics, bio-mathematics), be highly self-motivated, and able to work in a dynamic team. We expect the candidates to have a very good knowledge of statistics (preferably in the R environment), biodiversity modelling and diversity metrics at multiple scales. Having worked on those issues during the Master is a plus. Working language is English.
Knowledge of French is recommended but not mandatory.
Applicants will have to send their CV, last diploma, and a short presentation of their scientific project (2 to 3 pages max) to these email address: [email protected]; [email protected] & [email protected]
Letters of recommendation are welcome
Application deadline: June 15th 2017 at 17:00 (CET)
Applications will be evaluated through a three-step process:
1. Eligibility check of applications by June 15th 2017
2. First round of selection: the application will be evaluated by a Review Board in June 2017. Results will be given by June 2017.
3. Second round of selection: shortlisted candidates will be invited for an interview session in Grenoble by June-July 2017. (if necessary)
TYPE of CONTRACT: temporary-3 years of doctoral contract
JOB STATUS: Full time HOURS PER WEEK: 35
OFFER STARTING DATE: October 1, 2017
APPLICATION DEADLINE: June 15, 2017
Salary: between 1769 € and 1989.80€ brut per month (depending on complementary activity or not)