Satellite oceanography

Duration

15h Th, 15h Pr

Number of credits

Lecturer

Yves Cornet

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

Learning unit contents

As this course is geared towards students with very different technical and scientific backgrounds, its goal is to provide a common theoretical ground of general concepts used in the processing of digital images recorded by satellite sensors. We have decided to concentrate our formation program on these aspects because satellite oceanography is a very broad field of study, dealing with the observation of water bodies using sensors that detect visible light, reflective infrared, thermal infrared, and microwaves (hyperfrequencies - RADAR imagers). These sensors can be atmospheric sounder, radar altimeter; imaging instrument...
The atmospheric sounders are often used to modelize weather conditions and radiative transfer through the atmosphere. Processing data from such atmospheric sensors requires advanced knowledge in atmospheric and aerosol physics. Another field of satellite oceanography aims at determining a geoid model and measuring the sea surface height above/below this reference, while filtering disruptive effects (waves, tides ...). This is achieved by applying advanced concepts of geodesy, analyzing the satellites trajectories using their height measurements as provided by altimeters (e.g. RADAR altimeters) and their position and attitude provided by orbital positioning systems (e.g. DORIS) and by inertial measurement units (IMU) or star trackers. A third advanced topic of satellite oceanography is the processing of data acquired by SAR systems. Those systems which provide a phase and amplitude are used to get information on sea waves, sea surface state, and wind velocity ... The SAR processing technics call upon complex theoretical notions of signal processing.
This is why, for the introductory course, we have chosen to only study the data produced by imaging sensors that detect visible and infrared light. As this course is aimed at oceanographers and limnologists, it will be illustrated by examples that are specifically related to these fields of research. In addition, as the data acquired are geo-localized (geographical or spatial) and temporal, we will also explain general concepts of digital and mathematical cartography and spatial analysis, which are essential in order to analyze those data.
Whether the images are used to observe land or water bodies (lakes, seas or oceans), and whether the phenomena studied are physical, biological or anthropic, it is essential that students learn the general theory of image processing, regardless of characteristic dimension and time. This is what the theoretical part of the course focuses on. Most of the theoretical concepts are applied through the use of software tools during the supervised practical part of the course.
The course's general outline is as follows :
I. Introduction
1. Definition
2. Brief history
3. Satellite movements
4. Nature of the signal
5. Some satellites and sensors
 
II. Monogenic image processing
6. Concept of digital image
7. Monogenic image visualization
8. Contrast enhancement
9. Geometric corrections
10. Radiometric corrections
11. Spatial filtering
 
III. Polygenic image processing
12. Polygenic image visualization - color composites
13. Arithmetic operators and indices
14. Polygenic transformations
15. Image classification
 
IV. Examples of applications (informational part of the course)
16. Nature of satellite information in oceanography
17. Classification of the shallow water seabed
18. Bathymetry
19. Ocean color (OC)
20. Sea Surface and Lake Surface Water Temperature (SSWT and LSWT)
21. Sea surface height (SSH)
22. Radar imaging and of sea surface state
23. Front detection
24. Analysis of temporal series and teleconnections

Learning outcomes of the learning unit

* Given the diversity of the students attending this course, the requirements in terms of theoretical knowledge are not as high as could be expected from an expert who designs original solutions. Emphasis is rather placed on practical aspects. Still, students should follow basic scientific standards (rigor and reliability) and our expectations will obviously be tailored to each student's background.
* Understand the data acquisition process and the nature of the information recorded by imaging sensors used to observe water bodies.
* Understand the nature and the oceanological meaning of the data recorded by sensors studied in the course, as well as the different pre-processing levels of the images made available to scientists by the providers.
* Know the main types of processing used for these images.
* Master the functionalities of specific software tools allowing to apply these processes.
* Apply standard processing protocols and understand their limits and how they work.
* Understand why images are processed for oceanological purposes and interpret the oceanological meaning of the processes used

Prerequisite knowledge and skills

The course builds upon basic skills in mathematics, statistics, physics, spatial analysis and mathematical and numerical cartography. Students should also have an interest in computer science and programming.
Students will also be helped by the mindset they have acquired through various scientific courses (mathematics, statistics, physics, spatial analysis, etc.) and technical courses (numerical methods, programming, cartography, etc.) of former secondary education and academic programmes.
Nevertheless, the variety of backgrounds among the students who generally enroll in this course requires that the teaching be adapted. Many refreshers will thus be provided using the black board. In addition, students are given the opportunity at the beginning of each class to ask questions about the content from the previous class. It is therefore up to students to go through their notes every week in order to identify potential points of confusion.
The oceanographical interpretation of the results of the methods used in class will rely on knowledge of lake, sea and ocean hydrodynamics, oceanography and climatology... If necessary, students will search through the scientific literature in order to carry out this interpretation.

Planned learning activities and teaching methods

The theoretical part of the course is given as lectures. We also offer students an exercise book, featuring numerical examples illustrating the various methods studied during the lectures. Their goal is to help students understand the theoretical concepts that we have identified over the years as being the most difficult. Examples of typical solutions are provided. The exercises can be done using the computation or programming tools that are known to the students (Excel, Calc Openoffice, programming languages learned in IT class, scientific calculators, etc.).
The practical part of the course consists in assignments that are mostly completed using Idrisi sofware. It illustrates almost all the methods presented during the theoretical part of the course. Classes alternate between practical work and theoretical lectures. Students are also given standard exercises and datasets similar to those used in the practical assignments, along with the solutions, so that they can autonomously assess their ability to use the software before the exam.
Students are free to use the university's Idrisi license as well as other software applications through the ULg's VPN. For information on how to access these applications, students can visit the following web page: http://www.gitan.ulg.ac.be/cms.
This web site also features the timetable of the class room in building B5a (B5a/4/18 and B5a/2/35). To get access to these rooms, students may contact the Geomatics unit if they wish to practice or advance in their practical assignments.

Mode of delivery (face-to-face ; distance-learning)

The course consists mostly in face-to-face classes. Attendance is thus mandatory, but students can install open software on their laptops and use the ULg's licenses in order to progress as their own rhythm outside of class and the academic environment. Classes are held in room B5a/4/18 or B5a/2/35.

Recommended or required readings

MATHER P.M., 1999. Computer Processing of Remotely-Sensed Images. 2nd edition. Wiley, Chichester, 292 p.
RUSSELL G. CONGALTON & KASS GREEN, 2008. Assessing the Accuracy of Remotely Sensed Data: Principles and Practices. CRC Press, Second Edition.
Platform of Earth Observation (BELSO) : http://eo.belspo.be/ (viewed on 14/8/2014)
Landsat 7 handbook : http://landsathandbook.gsfc.nasa.gov/ (viewed on 14/8/2014)
Landsat 8 documentation: http://landsat.usgs.gov/landsat8.php (viewed on 14/8/2014)
Landsat Science : http://landsat.gsfc.nasa.gov/?page_id=11 (viewed on 14/8/2014)
NOAA documentation: http://www.ncdc.noaa.gov/oa/pod-guide/ncdc/docs/intro.htm (viewed on 14/8/2014)

Assessment methods and criteria

An ongoing non-certificational self-evaluation is carried out during practical sessions, through close interaction between students and teachers. In addition, students have a book featuring numerical exercises with solutions on the one hand, in order to assess their own theoretical knowledge. They also have access exam-type questions with solutions on the other hand, in order to test their skill in using the Idrisi software application.
The certificational evaluation will consist in an oral exam on the course's theoretical content and a problem to solve using Idrisi software, similar to those given during practical classes. Each part of the evaluation is worth 50% of the final mark.
This standard evaluation procedure may however be modified in agreement with the students, who will be notified of any change.
The assessment criteria are as follows: clarity, coherence, logic, rigorousness, precision, completeness, brevity, relevance, cross-cutting nature (within the course and between courses), quality of mathematical interpretations (mathematical meaning of the different coefficients of the equation, e.g.), physical interpretations (dimensions and units, order of magnitude - scaling, e.g.) and geographical and oceanographical interpretations (mono and multivariate spatial and temporal interaction and meaning - type - of the variables e.g.).
The critical mind with regard to the data used (qualification, nature, meaning, representativeness, standardization ...) and methodological choices (justification of the choice of methods, adopted thresholds ...) will also be taken into account in the evaluation.
Furthermore, answers will also be evaluated based on the quality and the originality of the graphic illustrations since graphic expression is the scientist's specificity. It further allows demonstrating a good understanding of the phenomenon. Finally, enriching an answer with a rich personal scientific culture will also be considered a factor of excellence in the assessment.

Work placement(s)

None

Organizational remarks

Classes are held according the planning of courses (http://www.facsc.ulg.ac.be/cms/c_253095/fr/horaires).  Theoretical lectures alternate with supervised practical work classes. Attendance is mandatory.

Contacts

Yves CORNET, Professor
Geomatics unit, 17 (B5a), Allée du 6 Août, 4000 Liège
Phone #: +32 4 366 53 71
E-mail: ycornet@ulg.ac.be
Web: http://139.165.44.35/cms/index.php

Items online