Outline content for SDM workshop
  Kuala Lumpur, 6-10 February 2011

This course will be a practical introduction to species distribution models (SDMs; also known as ecological niche models or bioclimatic models). SDMs predict species distributions by combining observed occurrence records with environmental data. They are used to develop ecological and evolutionary insights and to predict distributions across landscapes, sometimes requiring extrapolation in space and time.

This workshop will focus on both theoretical and practical aspects of SDMs, with particular emphasis on how to do a good job of fitting and evaluating a model.

Using a mixture of lectures, lab exercises, discussions, and case studies, participants will learn to:

  • identify relevant data, and prepare it ready for modelling;
  • fit models using at least two modelling methods, including MaxEnt;
  • evaluate models and interpret them;
  • understand the range of practical issues that arise in typical applications of SDMs.

Practical sessions will use free statistical software including R and MaxEnt. Example data will be provided but participants are also encouraged to bring their own data.

Topics covered in lectures:

  • Introduction to species modelling and typical applications
  • Ecological and geographic theory underpinning the models
  • Types and sources of species data, and particular challenges
  • Environmental data: what to aim for, where to find it
  • Modelling methods - overview
  • Modelling methods - quick discussion of regression, and machine learning techniques
  • Modelling methods - focus on Maxent and how it works
  • Understanding a model and evaluating it
  • Uncertainty
  • Challenges:
    • Predicting to new times and places
    • Spatial autocorrelation
    • Imperfect detection
    • Incorporating migration / dispersal into the models
  • Interesting case studies for discussion

Topics covered in practicals:

  • species modelling in R - introduction to, and practice with, package 'dismo'
  • modelling with Maxent - the basics, plus how to fine-tune models
  • modelling with boosted regression trees.
  • participant's data - group discussion

Back to BCSS Home Page


Page updated 12 Sept 2011 by Mike Meredith