Environmental Systems Analysis

ETH Zurich, Prof. Peter Reichert

(first time in the new format autumn semester 2007)

Goals:

•  Learn techniques to evaluate observed data with the aid of mathematical models.
•  Learn how to use mathematical models for predicting future behaviour of  environmental systems.
•  Get an overview of relevant numerical techniques and software and practice application of these techniques

Contents:

• Models in the environmental sciences:
  Importance of models, modelling uncertainty, mathematical representation of models
• Identification of models:
  Construction of models, estimation of model parameters (frequentist and Bayesian techniques, consideration of input uncertainty), sensitivity analysis, identifiability analysis (theoretical and practical approaches, Bayesian prior-posterior analyses), model structure improvement, selection and weighting (frequentist and Bayesian approaches), model testing
• Model predictions:
  Estimation of the uncertainty of model predictions.

Downloads:

Program 2007

R-Scripts:

simpdeg.r
sysanal.r

Dataset:

simpdeg.dat, simpdeg.new.dat

Exercises:

Introduction to the exercises
Exercise 1, R commands, Solution exercise 1, R Reference Card
Exercise 2, Solution exercise 2
Exercise 3, Solution exercise 3
Exercise 4, Solution exercise 4
Exercise 5, Solution exercise 5