Analysis of spatial and temporal data within geoscience

• list and describe commonly used covariance models
• describe the principle of maximum likelihood estimation
• explain the concept in a state-space model
• apply well known processes in modeling such as a random walk and an autoregressive model
• implement different algorithms for spatial modeling such as kriging and Gaussian Markov random fields
• compare and evaluate models
• discuss and defend choice of model
• design a spatial or temporal model to analyse real data
• document and present scientific work.

Spatial and temporal data are an integrated part of earth science. The content of this course will introduce some general tools to handle such data.
Some specific topics are:
• Maximum likelihood estimation
• Covariance functions
• Gaussian Markov random field
• State-space model
• Kriging
• Model evaluation

30201/30202/30230, Knowledge of programming

Lectures, exercises, project work in groups.