|Teaching method||Contact hours|
|Course coordinator(s)||dr. ir. VL Mulder|
|dr. ir. S de Bruin|
|Lecturer(s)||dr. ir. VL Mulder|
|dr. ir. S de Bruin|
|Examiner(s)||dr. ir. S de Bruin|
|dr. ir. VL Mulder|
Language of instruction:
Assumed knowledge on:
The course assumes basic GIS-skills (GRS-10306 Introduction Geo-information Science or GRS-10806 Geo-information Science for Planning and Design) and either GRS-20806 Geo-information Tools or SGL-30306 Inventory Techniques for Geosciences.
The provided overview of spatial and spatio-temporal modelling methods can be deepened in later courses such as GRS-30306 (Spatial Modelling and Statistics), deep learning courses and thesis research.
Representation and analysis of information on spatial phenomena and changes of these through time are key to many environmental disciplines concerned with soils, ecology, atmosphere, hydrology and the like. In this course, students extend previously acquired knowledge about spatial and spatio-temporal representations of information in the computer and use this knowledge on a selection of example applications in these domains. The course uses the R scripting language along with several contributed packages. It provides an overview of spatial and spatio-temporal analysis methods, including machine learning, which can be deepened in later courses such as GRS-30306 (Spatial Modelling and Statistics), a deep learning course and in thesis research. It is considered especially relevant for students in the fields of soil geography, geo-information science, landscape studies, hydrology and atmosphere or for students with an ecological background.
After successful completion of this course students are expected to be able to:
- explain contemporary concepts, methods and techniques from geo-information science used in spatial and spatio-temporal environmental analyses;
- apply machine-learning and interpret model outcomes in relation to environmental processes within the context of soil sciences;
- demonstrate the application of spatio-temporal representations from geo-information science using partially prepared computer scripts;
- evaluate the methods for spatial and spatio-temporal representation reported in scientific papers;
- design and carry out a study in which spatio-temporal concepts, methods and techniques are applied to an environmental problem.
The first week of the course focuses digital soil mapping using machine learning. This part of the course concerns 1) data handling and basic statistical analysis of spatial data, 2) developing a digital soil map using machine learning and 3) interpreting model metrics and maps in relation to environmental processes. The second week extends existing knowledge about spatial representation in Geographical Information Systems to the temporal domain. This part of the course concerns (1) movement of individual objects, which are studied using trajectories and space-time prisms, and (2) continuous space-time fields, with special attention for the value of sample information. In this course, theory is exercised using the R scripting language. In the third week students design and carry out a mini-project for a chosen application and discuss their results with fellow students and teachers. This project is reviewed and presented in week 4. The course closes with an open book exam.
- written exam that counts for 50% of the final mark, minimum grade 5.0 out of 10;
- the other 50% is gained from the practical assignments.
Journal papers and practical instructions are provided during the course.
|Restricted Optional for:||MEE||Earth and Environment||MSc||D: Soil Geography and Earth Surface Dynamics||4WD|
|MUE||Urban Environmental Management||MSc||4WD|