GRS-30306 Spatial Modelling and Statistics


Credits 6.00

Teaching methodContact hours
Practical extensively supervised51
Practical intensively supervised28
Course coordinator(s)dr. ir. S de Bruin
Lecturer(s)dr. ir. S de Bruin
dr. ir. GBM Heuvelink
dr. ir. A Ligtenberg
Examiner(s)dr. ir. S de Bruin
dr. ir. GBM Heuvelink

Language of instruction:


Assumed knowledge on:

Basic geo-information concepts and methods (for example GRS-10306 Introduction Geo-information Science); Basic statistics (for example MAT-14303 Basic Statistics; MAT-20306 Advanced Statistics.
Contents of chapters 3,4 and 11 of the book 'An Introduction to Statistical Methods and Data Analysis', sixth edition (Ott & Longnecker, 2006; ISBN: 0495109142).

Continuation courses:

SGL-, GRS-thesis (eg GRS-80436 MSc Thesis Geo-information Science and Remote Sensing).


Models of spatial processes are important means to support decision making. Such a model, however, is merely a formal and geo-referenced representation of a mental construction about the real world. This course offers knowledge and skills to qualify and implement spatial models and evaluate their results. The course consists of three modules: The first module provides an overview of spatial models and allows students to practice with a cellular automata-based model. The second part is about geo-statistics and includes practical's to generate spatial model input using sparse observations. The third module concerns the propagation of input errors through spatial models. A central case study based on a flood model is used as a connecting link throughout the course.

Learning outcomes:

After successful completion of this course students are expected to be able to:
- explain commonly used dynamic spatial modelling methods and techniques;
- construct a simple dynamic spatial model using Netlogo for a given spatial problem and data set;
- investigate the spatial correlation structure of a data set using semivariogram analysis;
- apply geo-statistical interpolation and spatial stochastic simulation methods; construct a stochastic error model to represent uncertainty in spatial data;
- apply fist order Taylor series and Monte Carlo methods for analysing propagation of input errors through spatial models.


- The topics are introduced in lectures and by reading scientific papers;
- practical's consist of step by step exercises related to each of the topics introduced in the modules;
- followed by more independent work on the central case study;
- Software tools to be used include ArcGIS, Netlogo and R (gstat Package).


- individual practical exercises (30% - 15% spatial modelling, 15% error propagation);
- an individual open book exam (70%).
The course guide gives details on the assessment procedure.


Literature and instructions will be provided.

Restricted Optional for: MGIGeo-Information ScienceMSc5MO