A group photo taken during the course
The “Ocean/Atmosphere Time Series Analysis: Theory and Practice II” course for oceanography, sponsored by CHESS, took place during the week May 8–12, 2017. The course was taught by Jonathan Lilly from NorthWest Research Associates in Seattle, and was held at the Forskningsparken at the University of Oslo.
The course covered classical and modern methods of time series analysis, with an emphasis on hands-on analysis of datasets of the students’ own choosing using Matlab, as well as on the underlying theory. Simple statistics and filtering, multi-taper spectral analysis, wavelet analysis, and the philosophy of data analysis were among the topics presented. The course ran from 10 AM until roughly 3 PM Monday through Friday, and consisted of about two hours of lectures and two hours of labs, followed by homework assignments for the evening.
The course was international in nature, with students from both Norway and Germany participating. Twenty-four students were in attendance, thirteen from Norway and eleven from Germany. These included graduate students from the University of Oslo (UiO), the University of Bergen (UiB), the University of Tromsø, (UiT), the Helmholtz Centre for Ocean Research Kiel (GEOMAR), and the Alfred Wegener Institute (AWI) in Bremerhaven.
At the end of the course, the students each produced a Matlab demo file documenting the application of each analysis method to their particular dataset.
The students and instructor all felt inspired at the end of a well-spent week. Students enjoyed the course and also felt it would be very beneficial to them in their future careers. As one student wrote, “My main aim for the course was to learn a strategy and different methods of analyzing a time series when I am confronted with one, so, a system of what to look for in the time series, in which order, etc. That was completely fulfilled and I am very happy about that.”
Written by Jonathan Lilly
Details about the course here.