Big Data in Sustainability Science

Course Objective
The key objectives of this course are:

to know how and when big data can be used to solve sustainability problems.
to gain a better understanding of methods and tools to analyze big data.
Course Content
Data is everywhere around us. But how can we make use of all this data to answer questions in sustainability sciences? How can twitter data help us to understand where and how severe people are affected by floods? Or how can we use earth observation data to classify land use?

In the course Big Data in Sustainability Sciences, you will learn the tools and knowledge to work with large datasets that are widely used within sustainability sciences. This ranges from filling gaps in census data through machine learning approaches, to working with earth observation data to map droughts.

The course is structured around six main data types and related methods. Specifically, we will:

explore census data and use machine-learning methods to fill data gaps and derive new information and knowledge;
better understand how you can use Earth Observation data to classify land cover.
assess how droughts affect us through using remote sensing data;
understand how to use open-source and Volunteered geographic information (VGI);
learn how to use social media data to identify natural hazard occurence and landscape preferences;
learn how the right visualization can help us to extract the right messages from our data.

In the first week, we will provide a crash course in Python that will be used throughout the remainder of the course. You are not expected to have any prior knowledge in Python before this course, but a clear interest in learning to code is recommended.

Additional Information Teaching Methods
This course will be a combination of lectures and tutorials. Each week
will consist of an introductory lecture to the method and/or data type
that will be applied and/or analyzed in that particular week. During the
introductory lecture, students will gain the required theoretical
knowledge to apply the methods during the tutorial. The tutorial each
week will be a half-day computer practical in which students will
develop the skills and knowledge to work with the method and/or
data-type through a hands-on assignment.
Method of Assessment
There will be six weekly assignments and a multiple-choice exam in the
final week. The weekly assignments will account for 60% of the grade
(10% per week), and the final exam will account for 40% of the grade.
The weekly assignments will be made in groups of two, whereas the final
exam will be individual.

Students must pass both elements (5.5 or higher).
Entry Requirements
There are no specific requirements, besides enrollment in the BSc Aarde,

Economie & Duurzaamheid.

Literature
Readings for each week are specified in the Canvas page. They consist of
book chapters and papers, and aim to give you a broad understanding of
the use of big data for sustainability sciences. It is very advisable to
familiarize with the readers before the lecture. This helps you to
actively engage in discussions during our meetings, to think of the
questions you want to ask the lecturer, and also makes it easier to
prepare for the final exam. The main notions of the readings can be
tested at the exam. Some items are specified as “background reading”,
and they are meant as suggestions to explore specific subjects in higher
detail. All readings are either freely available online from university
computers and from home by using the VU proxy server, or are uploaded to
Canvas.