Study Programmes 2015-2016
OCEA0045-1  
Statistical methods of analysis of oceanographic data
Duration :
20h Th, 10h Pr
Number of credits :
Master in space sciences (120 ECTS)3
Lecturer :
Aïda Alvera Azcarate
Language(s) of instruction :
English language
Organisation and examination :
Teaching in the first semester, review in January
Units courses prerequisite and corequisite :
Prerequisite or corequisite units are presented within each program
Course contents :
Oceanographic data, in-situ or satellite based, cover a wide time-space spectrum of processes, contain errors specific to each measurement system and require specific data analysis methods in function of the use one wants to make of them.
Learning outcomes of the course :
The lecture aims at describing the different measurement techniques used in hte marine environment and to introduce the different type of data analysis tools adapted to each type of data.


1.Statistical methods and error estimates : - probability, distributions, confidence intervals, regressions (linear, multivariate, correlation) - degrees of freedom, hypothesis testing - Obvious errors, rounding errors - interpolation - covariance 2. Spatial analysis - Objective analysis - Principal component analysis - Inverse methods 3. Temporal analysis - correlation function - Harmonic analysis - spectral analysis (FFT) - Wavelets - digital filters
Data acquisition and presentation will be also discussed during the lessons.
Prerequisite knowledge and skills :
A solid mathematical background and basic knowledge in ocean sciences
Planned learning activities and teaching methods :
Exercises will use real data sets from a CD-ROM. Matlab will be used for most exercises.
Mode of delivery (face-to-face ; distance-learning) :
Face-to-face learning
Recommended or required readings :
The main reference is
"Data analysis methods in physical oceanography", William J. Emery and Richard E. Thomson.
Other materials are available as well (see "on-line notes" section)
Assessment methods and criteria :
Skills will be assessed during exercise sessions using real inter-disciplinary data sets, taking into account the complexity and quality of methodology, processing, presentation, interpretation and synthesis.
Work placement(s) :
None
Organizational remarks :
Lectures will be given once per week, with sessions of 3-4 hours.
Interested students should contact me to establish the timetable of this lecture.
Contacts :
Aida Alvera Azcárate AGO-GHER Université de Liège Allée du 6 Août, 17, Bât. B5 4000 Liège, Belgium
A.Alvera@ulg.ac.be
Tel.: +32 (0)4 366 3664 Fax.: +32 (0)4 366 9729