Description: This module focuses on the statistical modelling of ecological data. It develops students’ ability to construct, fit, and interpret
statistical models that represent ecological processes, while explicitly accounting for uncertainty. Topics range from basic statistical concepts to linear mixed-effects models, Bayesian hierarchical models, Monte Carlo Markov chain methods, inverse modelling, time series, and state-space models. Case studies will focus on analysing plant and animal distribution, abundance, and demographic data while fully accounting for model, observation, driver, and process uncertainty. The module will use a combination of lectures, case study analyses, practical exercises and independent project work.
Learning outcomes:
Explain the principles underlying statistical modelling of ecological data.
Construct and fit linear, mixed-effects, and Bayesian hierarchical models to ecological datasets.
Implement MCMC and related computational methods for model inference and uncertainty estimation.
Apply time series and state-space modelling approaches to analyse temporal ecological data.
Evaluate and interpret model results, explicitly distinguishing among different sources of uncertainty.
Communicate and justify model assumptions, structure, and outcomes in an ecological forecasting context.
Module group: Foundations of Ecological Forecasting
Credits: 5 ECTS
Lecturers: Prof. Dr. Steven Higgins