SIO 221B:  Analysis of Physical Oceanographic Data

Winter 2026

SIOC 221B covers techniques for analysis of physical oceanographic data involving many simultaneous processes, including probability densities, sampling errors, spectral analysis, empirical orthogonal functions, correlation, linear estimation, objective mapping. It is the second course in a series. Students who have not taken the first course may want to look at the brief synopsis of key background that you should have at this web site.

Sarah Gille

Lectures: Wednesday/Friday 2:00-3:20, Spiess Hall 330

Problem session: Monday 2:00-2:50, Spiess Hall 330

SIO Office: Nierenberg Hall 407

Telephone: 822-4425 e-mail: sgille at ucsd.edu

Course website: http://pordlabs.ucsd.edu/sgille/sioc221b

Grading: S/U, based on homework and independent project. (See syllabus)

Final presentations: Monday, March 16, 3-6 pm

syllabus in pdf form
Reference List

This is the public website for the course. More information is available on the UCSD Canvas site for the course. When possible, schedule and topics will be posted here. Notes and topics for upcoming lectures are tentative and subject to extensive revision.

Lecture notes and handouts:   (See Canvas for slides, since they may contain copyrighted material.)

• Wednesday, January 7. Introduction to the course. Motivating a statistical framework. notes
• Friday, January 9. Probability density functions as a foundation. notes
• Wednesday, January 14. Joint probability density functions. notes
• Friday, January 16. Correlation, covariance, and random walks notes
• Wednesday, January 21. Random walks, diffusivity, and sampling error. notes
• Friday, January 23. Autocovariance and decorrelation time scales. notes
• Wednesday, January 28. Models and data: Least-squares fitting. notes
• Friday, January 30. Implementing least-squares fitting: Examples notes
• Wednesday, February 4. Uncertainties in least-squares solutions notes
• Friday, February 6. Weighted and constrained least squares. notes
• recorded lecture. Linear algebra review notes
• Wednesday, February 11. Eigenmodes and modal decomposition notes
• Friday, February 13. Singular value decomposition. notes
• Wednesday, February 18. More on empirical orthogonal functions and generalized inverse. notes
• Friday, February 20. no class
• Wednesday, February 25. Guest lecture, Luna Bai, Introduction to Machine Learning
• Friday, February 27. no class
• Monday, March 2. Linear estimation theory. notes
• Wednesday, March 4. Linear estimation theory applied to the ocean. notes
• Friday, March 6. Objective mapping examples for gradients and velocity. notes
• Wednesday, March 11. Guest lecture, Steve Diggs, University of California Office of the President, Data management for the future
• Friday, March 13. From objective mapping to machine learning, and connecting course themes and thinking about data assimilation notes; , barebones python example

Pitfalls people encounter in Matlab