Geodata Fundamentals: Applied Statistics with Python

Geodata Fundamentals brings together knowledge of geophysics and remote sensing, of statistics, and of computer science to lay the foundations for the learning line Earth System Data. Through practical examples, this course will teach you the basic principles behind making informed decisions under uncertainty to manage subsurface resources, mitigate geohazards, or predict future weather and climate, among other applications. 

Building around the concept of the data-to-insight workflow, you will learn the essential ingredients to acquire, manage, process, analyze, and interpret Earth system data based on two main parts: 

1. In the first part (Sensing and Statistics), you will focus on the principles behind Earth system data acquisition and analysis. Uncertainty is inherent to studying and managing Earth systems, so you will start with an introduction to important concepts in probability and statistics. Then you will learn about Earth system data acquisition using a selection of geophysical and remote sensing techniques.

2. In this second part (Applied Statistics with Python), you will focus on the practical use of statistics to start extracting meaningful insights from Earth system data using Python. Here you will learn to clean, transform, and analyze data and to build statistical models to predict properties related to Earth systems. You will work on case studies using the data you acquired yourself during the first part and data from other sources.

At the end of this course, you will have gone through the entire data-to-insight workflow based on an initial selection of data acquisition and analysis techniques. In the following courses of the learning line Earth System Data, you will progressively build your knowledge towards more specialized or advanced techniques, for instance to tackle the specificity of time series and spatial data, which are ubiquitous around Earth systems.

Formative assessment: 

This course includes three forms of formative assessment: 

• Multiple-choice and open questions during the interactive sessions, whose solutions are given and discussed in class. 

• Jupyter notebook assignments during the interactive sessions and practicals, which are discussed in class and whose solutions will be provided afterwards.

• Feedback on progress during the capstone case study from the teachers. 

Summative assessment: 

This part B of the course ends with a capstone case study consisting of an open question to be answered by applying the data-to-insight workflow using Python. The final grade is based on:

• A Jupyter notebook with your answer to the question and the analysis to support that answer (in group, 80% of the grade). 

• An exam consisting of open questions about the case study and your personal contribution (individual, 20% of the grade). 

A minimum grade of 5.8 is required to pass this part. In case of a fail, a repair option will be offered, which will have to be finalized at the end of week 5.2 and will lead to a final grade of 6.0 if sufficient.