Organizer: Vinciane Mossion

Flyer

The aim of this course is to introduce the underlying concepts and theories behind niche modelling and Species Distribution Models (SDMs) and their applications as predictive or explanatory tools in ecology. In particular, possibilities and limitations of SDMs will be reviewed (e.g. transferability in space and time) with case studies.

The practical session will allow students being able to:

• Calibrate different techniques of SDMs (from parametric to non-parametric models)
• Build and select the appropriate set of predictors
• Properly evaluate the model performance (fit and predictive power)
• Analyse the model structure (e.g. variable contributions, variance partitioning)
• Project the models in space and time, in particular under climate change scenarios

• Christophe Randin, University of Lausanne

The course will provide a comprehensive vision of:

• The relationship between SDMs and ecological niche theories
• Existing modeling techniques:
• Environmental predictor selection
• Model evaluation (e.g. metrics, tranferability with hindcasting)

For the practical, a dataset with a large amount of vascular plant species in the iconic Zermatt region will be provided. Each participant could work on a species of her/his choice to understand each step of modeling, from predictor selection and calibration to evalution and projection.

Program (updated 25.01.2018)

Each participant should bring her/his own laptop with the latest version of R installed, ideally R studio. If you need to, we can make a laptop available to you. Please contact Sara Santi for this.

Ideally, the participants have already participated in "An Introduction to R" training and have good skills in R and R language and good background in ecology.

Evaluation:

Practical will be evaluated by a short article of 2-4 pages (to provide 2 weeks after the end of the course) presenting the results of the modelling process on the Zermatt case study. An example essay can be dowloaded here. Note that you also have to include a section describing the variable selection procedure.

Date: 15-14 May 2018

Schedule: 8.55-18.00

Venue:  University of Neuchâtel, Faculté des Sciences, Emile-Argand 11, UniMail, building D, room D301-303

ECTS: 1.0 (Research tools)

Evaluation: Full attendance and active participation is required during the class. Practical will be evaluated by a short article of 2-4 pages (to provide 2 weeks after the end of the course) presenting the results of the modelling process on the Zermatt case study.

Information: Please contact Vinciane Mossion or the doctoral program coordinator Dr Sara Santi

Registration fee: free.

Meals: collective lunches are offered (cafeteria UniMail) and you are warmly invited to participate. No individual meal expenses will be reimbursed.

Travel expenses: For participants of the Interuniversity doctoral program in organismal biology (DP-biol ): see reimbursement conditions

Make sure that you sign the attendance list each and evey day and that you take your certificate of attendance at the end of the course (no attestation will be sent by mail)

• This course is free and open to all PhD students, however until 20 April priority is given to "Interuniversity doctoral program in organismal biology" participants.
• Post-docs are welcome as long as places are available.
• Maximum number of participants: 15 people (minimum 5 people).

Registration through the web only: Closed

Please note the cancellation policy (CHF 20) applicable if registration is cancelled after the deadline.

Deadline: 27 April 2018