Computational Data Analysis for Ocean and Earth Scientists

Learning Outcomes

Learning Outcomes

Having successfully completed this module you will be able to:

• Implement mathematical algorithms in Python to process and visualise data
• Derive and present meaningful conclusions from data analysis through quantifiable interpretation
• Select and justify the most appropriate statistical and signal-processing techniques to analyse data and answer different research questions

The module will introduce statistical analysis, curve fitting and the interpolation of data. The analysis of data in the frequency domain using the Fourier Transform will be covered with applications to filtering in 1-D and 2-D. The fundamentals of computer programming will be taught in practical sessions using Python and will involve implementing the techniques covered in the lectures. The course will include optimal methods for the display of data. The following is a week-by-week overview of the content covered.

Week 1: Introduction to the Module and Fundamentals of Python Programming

Week 2: Programming Loops and Functions

Week 3: Statistical Distribution and Correlation

Week 4: Regression and Model Selection

Week 5: Advanced Regression Techniques and Principal Component Analysis

Week 7: Finding Frequencies in Data

Week 8: Fourier Transformation, Energy and Power of Signal

Week 9: Filtering, Aliasing, Spectral Leakage, and 2D Fourier Transform

Week 10: Correlation, Cross Spectral, and Advanced Signal Analyses

Teaching and learning methods

Formal lectures: Lectures will systematically cover the main concepts and topics using PowerPoint presentations and Jupyter Notebooks. Illustrated handout materials will complement most lectures. Where relevant, lecturers' own research experience in the appropriate fields is brought into the lecturing sessions. References to the applicable chapter of course materials and/or relevant journal articles are provided to complement some of the lectures.

Practical sessions: Will exemplify the theory. Practical sessions will be computer-based exercises used to illustrate the concepts covered in the formal lectures. Computer practical sessions will use Python through Jupyter Notebooks.

A wide range of support can be provided for those students who have further or specific learning and teaching needs.

Assessment strategy

Assignment 1 (40%): You will write Python programs to analyse and plot Ocean and Earth science datasets using statistical methods learned. You will write short summaries of your findings and interpretations. Tests Learning Outcomes: 1, 2, and 3.

Assignment 2 (60%): You will write a substantial Python program, or a number of smaller programs, to process and analyse one or more datasets using the signal processing techniques learned. You will interpret the results of the analysis and present these as short reports including the analysis methods you have applied to the data. Tests Learning Outcomes: 1, 2, and 3.

Summative

This is how we’ll formally assess what you have learned in this module.